Louisiana State University LSU Digital Commons

LSU Historical Dissertations and Theses Graduate School

1986 Palladium-Catalyzed Vinylic Substitution of Aryl Halides on Polymeric Nitrogen Supports. Chia-hsing Sun Louisiana State University and Agricultural & Mechanical College

Follow this and additional works at: https://digitalcommons.lsu.edu/gradschool_disstheses

Recommended Citation Sun, Chia-hsing, "Palladium-Catalyzed Vinylic Substitution of Aryl Halides on Polymeric Nitrogen Supports." (1986). LSU Historical Dissertations and Theses. 4268. https://digitalcommons.lsu.edu/gradschool_disstheses/4268

This Dissertation is brought to you for free and open access by the Graduate School at LSU Digital Commons. It has been accepted for inclusion in LSU Historical Dissertations and Theses by an authorized administrator of LSU Digital Commons. For more information, please contact [email protected]. INFORMATION TO USERS

This reproduction was made from a copy of a manuscript sent to us for publication and microfilming. While the most advanced technology has been used to pho­ tograph and reproduce this manuscript, the quality of the reproduction is heavily dependent upon the quality of the material submitted. Pages in any manuscript may have indistinct print. In all cases the best available copy has been filmed.

The following explanation of techniques is provided to help clarify notations which may appear on this reproduction.

1. Manuscripts may not always be complete. When it is not possible to obtain missing pages, a note appears to Indicate this.

2. When copyrighted materials are removed from the manuscript, a note ap­ pears to indicate this.

3. Oversize materials (maps, drawings, and charts) are photographed by sec­ tioning the original, beginning at the upper left hand comer and continu­ ing from left to right in equal sections with small overlaps. Each oversize page is also filmed as one exposure and is available, for an additional charge, as a standard 35mm slide or in black and white paper format. *

4. Most photographs reproduce acceptably on positive microfilm or micro­ fiche but lack clarify on xerographic copies made from the microfilm. Fbr an additional charge, all photographs are available in black and white standard 35mm slide format.*

*For more information about black and white slides or enlarged paper reproductions, please contact the Dissertations Customer Services Department

T T A /fT Dissertation L J 1V 1 1 Information Service

University Microfilms International A Bell & Howell Information Company 300 N. Zeeb Road, Ann Arbor, Michigan 48106 8629202

Sun, Chia-Hsing

PALLADIUM-CATALYZED VINYLIC SUBSTITUTION OF ARYL HALIDES ON POLYMERIC NITROGEN SUPPORTS

The Louisiana State University and Agricultural and Mechanical Ph.D. Col. 1986

University Microfilms

International300 N. Zeeb Road, Ann Arbor, Mt 48106 PLEASE NOTE:

In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified herewith a check mark -/

1. Glossy photographs or pages_____

2. Colored illustrations, paper or print______

3. Photographs with dark background____

4. Illustrations are poor copy______

5. Pages with black marks, not original copy______

6. Print shows through as there is text on both sides of page______

7. Indistinct, broken or small print on several pages _

8. Print exceeds margin requirements_____

9. Tightly bound copy with print lost in spine______

10. Computer printout pages with indistinct______print

11. Page(s)______lacking when material received, and not available from school or author.

12. Page(s)______seem to be missing in numbering only as text follows.

13. Two pages numbered . Text follows.

14. Curling and wrinkled pages______

15. Dissertation contains pages with print at a slant, filmed as received______

16. Other______

University Microfilms International Palladium-Catalyzed Vinylic Substitution of Aryl Halides on Polymeric Nitrogen Supports

A Dissertation Submitted to the Graduate Faculty of the Louisiana State University and Agricultural and Mechanical College in partial fulfillment of the requirements for the degree of Doctor of Philosophy in The Department of Chemistry

by Chia-Hsing Sun B.S., Tunghai University, 1976 M.S., Tsinhua University, 1980 May 1986 ACKNOWLEDGEMENT

The firm instruction and guidance by Dr. William H. Daly, the director of this research is sincerely appreciated. However, the work could not have been completed without the constant spiritual support and instruction of the author's parents, to whom this work is dedicated. The author further appreciates sharing discussions and experiences with Mr. Amnard Sittatrukul. Financial support by Louisiana State University was provided for the author throughout the course of his research. The author gratefully acknowledges financial assistance from the Dr. Charles E. Coates Memorial Fund of the Louisiana State University Foundation for expenses incurred in the publication of this Dissertation.

i Table of Contents Page ACKNOWLEDGEMENT...... i LIST OF T A B L E S ...... vii LIST OF ILLUSTRATIONS...... '. . ix LIST OF FIGURES...... xi ABSTRACT...... xil I. INTRODUCTION...... 1 II. PALLADIUM-CATALYZED REACTIONS ...... 5 2.1 Introduction and Background. .... 5 2.2 Chemistry of Palladium Compounds (Inorganic and Organometallic

Aspects) ...... 7 2.2.1 Complexes of Palladium(o). . 7 2.2.2 Properties of complexes of (PR3)ilP d ...... 10

2.2.3 Reactions of Palladium(o) complexes...... 11 2.2.4 [Pd(dba)g]...... 13

2.2.5 Complexes of Pd ( I ) ...... 14 2.2.6 Complexes of Pd(II)...... 15 2.3 Application of Organopalladium Catalysiss in Homogeneous Catalysis in Organic Synthesis...... 18 2.3.1 Arylation of Alkenes .... 22 ii 2.3.2 Vinylation of Alkenes. . . . 40 2.3.3 Palladium-catalyzed Nucelo- philic Additions to Alkenes. 49 III. POLYMER SUPPORTED CATALYSIS...... 62 3.1 Introduction...... 62 3.2 Synthesis of Polymer-Supported Catalysts ...... 68 3.2.1 Functionalization of Preformed Polymers ...... 71 3.2.2 Polymerization of Monomeric Ligands...... 81

3.2.3 Polymerization of Monomeric Organometallic Complexes . . 86 3.2.4 Macromolecular Immobili­ zation ...... 89 3.2.5 "Single Crystal" Polymer Support...... 91 3.3 Reaction Aspects of Polymer- Supported Catalysts...... 91 IV. RESULT AND DISCUSSION...... 104 4.1 Synthesis of Dipyridyl Derivatives for Immobilization on Polymer . . . 104 4.2 Synthesis of Pyridine Derivatives

for Immobilization on a Polymer Support...... 111 iii 4.3 Homogeneous Palladium-Catalyzed Vinylic Substitution...... 115 4.4 Substituent and Concentration Effect of Pyridine and Dipyridyl. 122 4.5 Substituent Effect of Aryl Halides...... 130 4.6 Polymer-Supported Catalyst. . . . 140 4.7 Conclusion...... 152 V. EXPERIMENTAL SECTION ...... 154 General Information...... 154 Specific Reaction Procedures ...... 155 1 . 2-(2-pyridyl)cinchoninic acid . . 155 2. Ethyl 2-(2-pyridyl)cinchoninate Method A ...... 156 Method B ...... 156 3. N-(3-propylamino)-2-(2-pyridyl)- cinchoninamide...... 157 4. Benzyl 2-(2-pyridyl)cinchoninate. 158

5. 4-pyridine carboxylic acid. . . . 159 6. Benzyl 4-pyridinecarboxylate. . . 159 7. 4-Benzyloxypyridine Method A (attempted)...... 160 Method B ...... 161 Method C ...... 161 8. 4-Hydroxypyridlne-1-oxide .... 162 9. 3-Benzyloxypyridine ...... 163 iv 10. 4-Benzyloxypyridine-1-oxide . . . 163 11. Vinylbenzyl acetate ...... 164 12. 4-Vinylbenzyloxypyridine-1-oxide (attempted)...... 164

13. 4-Vinylbenzyloxypyridine...... 165 14. Vinylbenzyl 2-(2-pyridyl)- cinchoninate...... 166 15. Vinylbenzyl 4-pyridinecarboxylate 16. PdC0Ac)2...... 168 17. Terephthaldehyde...... 168 18. p-Bromo benzylalcohol ...... 169 19. 4-Vinylbenzaldehyde (attempted) . 170 20. Dibenzalacetone (DBA) ...... 171 21. Pd2(DBA)3CHCl3 ...... 171 22. 15? DVB cross-linked [poly- vinylpyridine)] ...... 172 23. DVB cross-linked [poly- (vinylbenzyloxy-4-pyridine)]. . . 173 24. DVB cross-linked [poly- (vinylbenzyl-4-pyridine carboxylate) ] ...... 173 25. DVB cross-linked [poly- (vinylbenzyl 2-(2-pyridyl)

cinchoninate) ]...... 174 26. 15? DVB cross-linked [poly- (vinylpyridine) PdCl2)...... 174 v 27. [PolyCvinylbenzyloxy-4-pyridine) PdClg 3...... 175 28. DVB cross-linked Epoly- Cvinylbenzyl 4-pyridine-

carboxylate) PdCl23 ...... 175 29. General vinylic substitution procedure for homogeneous catalysis...... 175 30. General vinylic substitution procedure for heterogeneous

* catalysis...... 176 BIBLIOGRAPHY ...... 179 APPENDIX Attempted synthesis of p-vinylbenzyl chloride...... 201 VITA ...... 209

vi LIST OF TABLES NUMBER Page

I. Regiochemistry of Phenylation...... 26 II. The relative rates of phosphonium salt formation...... 32 III. Influence of substituted triarylphosphine ligands on vinylic substitution of methyl acrylate with 4-bromophenol...... 33 IV. Aryl halides exhibiting limited

reactivity with methyl acrylate...... 34 V. Ratio of terminal to internal addition

to 1-hexene...... 43

VI. Comparison of homogeneous and hetero­ geneous catalysts...... 64 VII. Catalyst types employed by chemical industry...... 65 VIII. Comparison of inorganic and organic supports...... 67 IX. Relative rates of hydrogenation of various olefinic substrates by homo­ geneous and anchored Wilkinson's catalyst...... 93 X. Associated effect of ligand, base and solvent on the reactivity and solubility of Pd complex...... 118 vii XI. Effect of pyridines on the reactivity of vinylic substitution ...... 129 XIa. Change in slope of initial rate vs. ligand concentration ...... 131 XII. Substituent effect of aryl iodides . . . 131 XIII. Substituent effect of aryl bromides. . . 134 XIV. Isolated styrene derivatives ...... 138 XV. Nitrogen-containing monomers ...... 140 XVI. Polymer-supported catalysts...... 142 XVII. Catalytic activity of polymer-supported

catalyst...... 145 XVIII. Solvent effect of xylene ...... 149 XIX. Metal leaching study ...... 151

viii LIST OF ILLUSTRATIONS SCHEME NUMBER Page I. Major reactions of cr-organopalladium

complexes...... 19 II. Nucleophilic addition of Pd-olefin complexes...... 21 III. Precursors, of arylpalladium intermediates...... 21 IV. Mechanism of Heck type reaction...... 23 V. Mechanism of arylation of 3-buten-2-ol . 36 VI. Precursors of vinylation intermediates . 40 VII. Vinylation of methyl acrylate...... 42 VIII.

complex...... 57 XIII. Derivatives of additions to olefins. . . 59 XIV. Regio- and sterochemistry of addition

of olefin...... 59 XV. Dissociation of Pd-N complexes...... 122 XVI. Replacement of Cl on Merrifield resin. . 72 XVII. Complexation of Cp on Merrifield resin . 73 ix XVIII. Complexation of polymeric phosphine. . . 75 XVIX. Carboxylation and sulfonylation of 2,2'- dipyridyl...... 105 XX. Derivatization of 2-(2-pyridyl)- cinchoninic a c i d ...... 107 XXI. Derivatization of 2-(2-pyridyl)- cinchoninoyl chloride...... 108 XXII. 4-Pyridinecarboxylic acid...... 111 XXIII. Benzylation of 4-pyridinecarboxylic a c i d ...... 112 XXIV. Benzylation of 4-hydroxypyridine .... 113 XXV. 4-Benzylamino Pyridine ...... 115 XXVI. Life cycle of Pd catalyst...... 121 XXVII. Mechanism of catalytic reaction of Pd-N

complexes...... 127 XXIX. R.D.S. of palladium catalysis...... 136 XXX. Attempted preparation of p-vinylbenzyl chloride for terephthaldehyde...... 206

x LIST OF FIGURES NUMBER Page I. Palladium-catalyzed vinylation of aryl halides...... 24 II. Orbital overlap of Pd-olefin complex . . 60 III. Typical plot of reaction rate, i.e. disappearance of styrene or aryl halide vs. time...... 123 IV. Dependence of catalytic reactivity on the ratios of 4-picoline vs. Pd...... 124 V. Dependence of catalytic reactivity on the ratio of 2,2'-dipyridyl vs. Pd . . . 126 VI. Catalytic activity of Pd-dipyridyl complexes...... 132

VII. Hammett P l o t ...... 135

xi ABSTRACT The structure of typical palladium complexes used as catalysts in organic synthesis is reviewed; particular attention is paid to the systematic study of palladium catalyzed vinylic substitution on a variety of aryl iodides and bromides (Heck-type vinylation). The general conditions of this process, i.e., a homogeneous catalyst operating at moderate temperature, which requires a proton acceptor to complete the catalyst cycle, are conducive to successful development of a polymer-bound catalyst. Using 4-picoline as a ligand under homogeneous Heck-type vinylation conditions. Morpholine is an adequate base to stabilize the catalytic picoline complex when aryl iodides are condensed with styrene. However, sodium bicarbonate in the presence of morpholine is required to effect the coupling reactions when aryl bromides are employed as the substrate. The following amines have been evaluated as ligands: 2-picoline, 4-picoline, 2-aminopyridine, benzyl 4-picolinate, 2-methoxypyridine, 3-chloropyridine and 2,2'-bipyridyl. Complexes derived from benzyl 4-picolinate exhibit higher catalytic activity than those derived from 4-picoline. Changing the ligand-to-palladium molar ratio from 1:1 to 3:1 for the bidentate ligands, bipyridyl and benzyl 2-(2-pyridyl) cinchonate, reduces catalytic activity two-fold.

xii Transformation of the above mentioned catalyst systems into polymer-bound catalysts entailed synthesis and polymerization of the following monomers: 4 r(31)-vinylbenzyl-4-oxypyridine, 4'(31)-vinylbenzyl-4-pyridinecarboxylate, and 4* C3’)-vinylbenzyl-2-(2-pyridyl)cinchoninate. 4-vinylpyridine and each of the vinylbenzyl monomers was copolymerized with divinylbenzene to produce cross-linked, insoluble resins with pendant ligand contents ranging from 1.57 to 8.27 meq/g. Treatment of the polymeric amine resins with sodium tetrachloropalladate produced polymer-bound palladium complexes capable of promoting the condensation of aryl halides with styrene. The initial rates of Heck type vinylation on the first cycle are nearly identical to those observed in the presence of a soluble catalyst complex. Thus, reactions in the presence of polymer-bound, heterogenous catalysts are not subject to diffusional limitations. A significant decline in catalyytic activity is observed when the catalyst is recycled a second or third time. Loss of catalytic activity can be traced to palladium leaching by an essential co-solvent, DMF or HMPA, and to occlusion of palladium in the precipitated olefin products. xiii I. Introduction Palladium complexes are useful catalysts for forming carbon-carbon bonds. Palladium-catalyzed arylatlons of alkenes, also called palladium-catalyzed vinylation of aryl compounds, was discovered by 1 2 Mizoroki and Heck. The most typical approach for the Heck*s arylation employs aryl halides as the precursors of the arylpalladium species. A systematic study of palladium-catalyzed vinylic substitution on a variety of aryl iodides and bromides has been performed by Heck. He observed that the tris-(o-tolyl)phosphine ligand yielded the most efficient and stable palladium catalyst for the reaction. Further, he showed that a proton acceptor was required to complete the catalytic cycle. The general reaction is: H + RX + Base

R + BaseHX (6)

R = Aryl, Heterocyclic X = Bromide, Iodide or Chloride (rarely) L = Ligand (Ph^P is generally used)

1 2

The proton acceptor may be either a secondary or tertiary amine and/or sodium or potassium acetate, carbonate, or bicarbonate. The catalyst is commonly palladium acetate; palladium chloride or preformed tertiary phosphine palladium complexes have also been used. A reactant, product, or solvent may serve as the ligand L in reactions Involving organic iodides, but a tertiary phosphine or a secondary amine is required when organic bromides are used. To keep the reaction homogeneous, solvents such as acetonltrile, dimethylformamide(DMF), hexamethylphosphoramide(HMPA), N->methylpyrrolldlnone(NMP) and methanol have been used. The reaction occurs between 50°C and 160°C. Chalk reported that pyridine-palladium complexes were too stable to be catalytically active in the reaction of halobenzenes with methallyl alcohol. However, Heck reported that nitrogen containing ligands themselves were not effective in catalytic reaction in the absence ii of triphenylphosphine. Rapid precipitation of metallic palladium occurred soon after the addition of the aryl halide substrate. We postulated that binding the palladium metal to a polymer containing basic functional groups might facilitate the proton transfer step and enhance the activity of the catalyst. 3

Obviously, the immediate advantages of this polymer-supported catalyst are the ease of its recovery from reaction solution and the possibility of recycle. Another advantage is based on consideration of ligand stability in the airj triphenylphosphine and polymeric phosphines are readily oxidized, while pyridine and polymeric pyridine3 are relatively stable. Careful perusal of the literature did not reveal a definitive study of the influence exerted by different types of bases on the coupling reaction. Thus, we elected to evaluate the influence of base structure on palladium catalyst stability and activity in aryl substitution reactions. The research demonstrated that efficient proton scavengers stabilize amino-palladium complexes. The objective of the research is to develop and exploit a new polymer-bound palladium catalyst system, derived from poly(4-vinylpyridine). We have shown that a soluble catalyst derived from 4-picoline, a model compound for the polymer, is stable in the presence of non-nucleophilic bases like 1,8-(bis-N,N-dimethylamino)naphthalene (proton sponge). Other models for different types of polymer systems such as benzyl 4-picolinate, and benzyl 2-(2-pyridyl)cinchoninate have been synthesized. Using appropriate proton scavengers it is possible to prepare soluble catalysts from each of these model systems and to ascertain the ligand substituent effects on the reactivity and stability of the palladium complexes. We have established the general reaction conditions required to effect vinylic substitution. In addition to a proton sponge, a polar aprotic solvent is required; reaction with the more reactive aryl iodides occurs in dimethylformamide (DMF), but the presence of hexamethylphosphoramide (HHPA) is essential to effect reaction with aryl bromides. The following factors were varied to elucidate the reactive and mechanistic aspect of catalysis: a) concentration of ligand, b) substituent effects on the aryl iodides and bromides. Monomers of several pyridine derivatives and one dipyridyl derivative were prepared and polymerized as required, the polymeric ligand thus synthesized was then complexed with lithium tetrachloropalladate to generate the resulting polymer-supported catalyst. Comparison of the results from our homogeneous catalyst studies with those obtained with heterogeneous polymer-bound catalysts indicates that the immobilized catalysts promote coupling reactions at rates comparable to the homogeneous model systems. II. Palladium-Catalyzed Reactions 2,1 Introduction and Background Palladium is a relatively rare element occurring to the extent of one part in about 10 J of the Earth's crust. It is extracted from copper-nickel ores which are found in Canada, South Africa and Russia. Palladium is grey-white metal which is ductile and malleable. It has a high melting point C1551*°C> and boiling point (3980°C) and is very resistant to corrosion. Palladium falls in the nickel triad (Ni, Pd, Pt) of the group VIII elements with d^° electronic configurations which readily undergo PdCO) Pd(II) redox interconversions. The facile redox couple enabled J. Smidt and coworkers to develop the well-known palladium-catalyzed Wacker process for the industrial c production of acetaldehyde from ethylene and O g , and later vinyl acetate from ethylene, Og and acetic acid.®'^ Palladium salts or complexes have been used extensively as heterogeneous or homogeneous catalysts in a diverse variety of organic reactions. Such catalytic reactions include oxidative and nonoxidative coupling of substrates such as olefins, dienes, acetylenes, and aromatlcs; and various isomerization, 6

disproportionation, hydrogenation, dehydrogenation, carbonylation, arylation and vinylie substitution reactions, as well as reactions leading to the formation of bonds between carbon and halogen, nitrogen, sulfur and silicon... . Palladium compounds have been found in 0, +1, +2 and +4 oxidation states. Among them, PdClI) complexes are the most abundant species, mainly because of the instability of palladium complexes with the other three oxidation states. The Pd(0) compounds occur only when stabilized by certain types of ligands called Tt -acid ligands, e.g., CO and phosphines. PdCO) complexes with d^° electronic configuration possess very high electron densities and tends to collapse into bulk metal. JL -acid ligands can accept electron density from the metal.

M <- L

An important series of 71 -acid ligands in catalytic chemistry of the transition elements is based upon the 7

trivalent compounds of P, As and Sb. These compounds have empty d-orbitals of proper symmetry and energy to overlap with the filled metal d-orbitals.

2.2 Chemistry of Palladium Compounds: (Inorganic and Organometallic Aspects) 2.2.1. Complexes of Palladium(O), d 10 Metallic palladium has been the classic source of Pd(0). It is a general characteristic of transition metals that they can form compounds in the zero oxidation state. The properties of these zero valent compounds are quite different from those of crystalline metal. The bulk metal is quite unreactive while the zero valent compounds are very reactive. There are several ways to generate monomeric Pd atom and Pd(0) compounds which can keep Pd(0) in a monomeric atomic active form. Metallic palladium is activated to serve as a catalyst for hydrogenation by being dispersed on various supports such as carbon, barium sulfate, alumina, etc. A highly active form of metallic 8

palladium is produced upon reducing palladium chloride with potassium in the presence of triethylphosphine in D refluxing THF. (Ph^PJ^Pd, prepared by reducing a bisCtriphenylphosphine) palladium chloride salt with hydrazine in the presence of triphenylphosphine,^ is the most commonly used Pd(0) reagent (equation 2).

2 (Ph3P)2 PdCl2 + 4 Ph3P + 5 N2H4 — >

2 (PhgP)^ Pd + it N2h £cT + N2 (2)

1 0 Other reducing agents include carbon monoxide, metal alkoxide (e.g. sodium n-propoxide, potasssium t-butoxide, 11 12 13 sodium acetate and sodium hydroxide), * aluminum alkyls, ^3»15 e^harioj (12,16 an£j even benzylamine. ^

Because (Ph3P)^Pd undergoes slow decomposition when exposed to the air, coordinatively unsaturated palladium(O) complexes, looked upon as the active

species of (Ph3P)2Pd in solution, are generated in situ, by simply reducing of (Ph3P)2PdC12 with two equivalents of diisobutylaluminum hydride in THF. 21 Among phosphines, the most interesting and potentially useful complexes are bisE1,3-bis(diphenylphosphino) 18 ethane] palladium (dppe2pd) and bis[1,3-bisE1,3-bis(diphenylphosphino)propane] palladium (dppp2pd). 9

In addition to phosphines, dibenzylideneacetone (DBA) was successfully used as a ligand. Bis(DBA)palladium(O), synthesized by the treatment of DBA with Na2PdCljj in hot methanol, has found useful applications.^ BisCDBA)palladium(O), a dark purple needle, is fairly stable in air in the solid state, but slowly decomposes in solution to black metallic palladium and yellow DBA crystals. The complex is slightly soluble in CHgClgi CHC13 and benzene. The presence of Pd(0) is confirmed by the fact that, in the presence of excess Ph^P, (Ph^PJ^Pd is synthesized from [Pd (dba)2]j more correctly formulated as [Pd2(dba)3] dba. When this is recrystallized from benzene, chloroform or dichloromethane, the solvent complexes pn [Pdg(dba)^Csolv)] are isolated. The [PdgCdba^Csolv) ] complexes readily undergo oxidative addition to give a variety of products. Complexes with mixed CO and PhjP ligands have been prepared. For example, in the presence of CO and excess Ph^P, (Ph3P)2PdC12 is reduced with NaBH^ in methanol to give Pd(CO) (PPhg)^! Pd3(CO)3(PPh3)3 and Pd3(CO)3(PPh3)1^.^^'^^ Zero valent palladium carbonyls without phosphine ligands have been prepared by using the technique of matrix isolation at low temperatures. Pd(CO)n (n=1 to 4) 10

pii 2*5 have been characterized by infrared spectra. ’ The isocyanides, RNC, which are isoelectronic with CO, have been shown to stabilize both Ni(0) and Pd(O). The reduction of Pdl2 (CNR)2 in strongly alkaline solution2^*2^ gives the polymer [Pd(CNR)23n(R=Ph, P-meCgH^, P-meoCgH^). 2.2.2. Properties of complexes of phosphinepalladium(O): (PR^^Pd All the phosphine complexes (PR^^Pd tend to dissociate in solution by giving off PR^ group. The size (or cone angle) of phosphine decides the extent of dissociation. For dialkylaryl and trialkyl phosphines, dissociation does not occur. With PCCyclohexyl)^, which has a very large cone angle, Pt(PR3)2 can be p O isolated. Complexes of lower coordination number, 2Q 30 PdCPR^)^ n=2 or 3, have been prepared. * 31 P NMR has proved to be a powerful technique to investigate the phosphine complexes present in 31 solution. Using this technique, it was found that four-coordinate complexes undergo facile dissociation to give the coordinatively unsaturated species [Pd(PR3)3] (16 electrons) and [Pd(PR3)23 (14 electrons). This observation accounts for the relative ease at which [Pd(PR3)^] undergoes oxidative addition. The product L2 XYPd satisfies the 18-electron rule. The facile interconversion of 11

Pd CO) Pd(II) accounts for the outstanding catalytic properties of palladium compounds. 2.2.3. Reactions of palladium(O) complexes Phosphine palladium(O): Palladium(O) complexes undergo oxidative addition with diverse variety of alkyl halides and aryl halides. 32 *33

Pd(PPh3)1( + R X > trans-[Pd(PPh3)2 RX] (3)

The insertion of palladium(O) into carbon-halogen bonds has been applied to the synthesis of Pd(II) complexes. The efficiency of the insertion of Pd into the carbon-halogen bond of aryl halides is in the order C-I > C-Br > C-Cl. For aryl halides, a similar order I > Br > Cl is found for the reaction with Pd(0).^ Chlorobenzene was inactive under the conditions employed for iodobenzene and bromobenzene, however electron withdrawing groups of substituted aryl chlorides tend to enhance the reactivity. It was suggested that the mechanism of oxidative addition of PdCO) is presumably S^2 reaction in which the cleavage of carbon-halide bond is the rate-determining step as shown in equation 4. X = Cl, Br, I R = Substituted Functional Groups

For highly reactive alkyl halides, such as o< -bromoesters, benzylbromide and 2°-alkyl halides, reaction can, in some situations occur by one-electron radical pathway.

M R-X R-M-X

X-W-X + R R-M-X

For much less reactive alkyl chlorides, modification of reactivity is possible by simply functionalizing them with -CN, -NO^ or fluorinated 13 groups. The functionalized alkyl chlorides will undergo oxidative addition. 35 Other variation on the oxidative addition of [PdCPPhJg]^ with S02 ,37 CS2 , NO, 02 , polyfluorovinyl halides37,33 terminal acetylenes3^ 40 and main group organometallics, has been reported. 2.2.4. Dibenzylideneacetonepalladium(O) complexes: [Pd(dba)2l Ligands used for the preparation of palladium(O) complexes have been limited to phosphines, phosphites, arsines, isocyanides, and carbon monoxide. Among them, only PPhg is most frequently used to produce (PPh^nPd which is a very versatile palladium complex as shown previously. [Pd(dba>2], a novel and stable complex, forms an unstable adduct upon reaction with olefins. In the presence of a stabilizing ligand (bipy, o-phen, P(0Ph)g or P(OMe)^), the complexes [PdL2(olefin)] were isolated from benzene or acetone solutions. 41 * 42 * 43 J Olefins, containing electron withdrawing groups, e.g., maleic anhydride, dimethylmaleate, dimethylfumarate, divinylsulfone or acrylonitrile, were used. A range of allyl halides undergo oxidative addition to [Pd2 (dba)g(solv)] to form allyl palladium 44 dimer. The relative reactivity decreased in the order allyl bromide > allyl chloride > methally chloride > crotyl chloride > cinnamyl chloride. 2.2.5. Complexes of palladium(I): Very few Pd(I) complexes have been made, however there is an important chemistry in the +1 state, where g Pd-Pd bond is involved. Pd Cl) has a d electronic configuration and is predicted to be paramagnetic. On the contrary, almost all Pd(I) complexes are diamagnetic and at least dinuclear with Pd-Pd bond. Ligands which stabilize Pd(I) by bridging two metal atoms include allyl, cyclopentadienyl, benzene, bidentate ligands, isocyanites and carbonyls. Basically, Pd(I) complexes can be made easily and are quite stable in the solid. The main approach for generating dinuclear PdCD complexes is by disproportionation of Pd(II) complexes and Pd(0) complexes. Some examples of dinuclear Pd(I) complexes are as follows. (Equations 5 and 6.)

x«ci,aici4

,45 0*0 15

^ 7 ”pd\ Y ~ ^ + 2 PhjP >Ph3P— — ph3p (5)46 1 'l'

-Br THF (6)U7 P d's.i + ^ ^i-PTjP--- pd-- Pd--- Pi-Prj ^ P i-Pr3 \ I Br

A few examples of PdCD complexes with trinuclear hg (2) and even fewer withLth tetranuclear structures have been reported (3). 49

° ^ ? P\ 7 ° 2 LNpd- -//

L L L = Bu'NC u L „ L = PMePhj

2.2.6. Complexes of Palladium(II): Palladium(II) complexes are among the most used catalysts for the synthesis of both industrial and fine chemicals. Unfortunately, the most commonly used palladium(II) salt, palladium chloride, is virtually 16 insoluble in most organic solvents. Palladium(II) chloride exists in o( - and fS -forms. The unstable^ -form is a linear chain of double chloride-bridged PddD's.^0 The j? -form, commercially available, consists of octahedral clusters of six Pd(II) atoms which are cross-linked by chloride bridges and it is Cl more stable. Palladium chloride is readily converted into a soluble NagPdCl^ salt by treatment with two equivalents of NaCl in situ. 52

PdClgCPhCN^ and PdC^tMeCN)^ complexes are formed simply by dissolving the PdCl2 salt in PhCN or MeCN, respectively, followed by evaporation. Palladium acetate was made by heating palladium sponge in a mixture of nitric acid and acetic acid. It is soluble in benzene and alcohols.

(PhCNjgPdCl^, readily prepared from benzonitrile and palladium chloride, is a convenient starting material for the preparation of olefin derivatives. The wweakly-bonded benzonitrile is apt to dissociate in organic solvents allowing easy complexation of olefins to PdCl^. Palladium(II) salts form many complexes with nitrogen-containing ligands which have square planar geometry. The simplest example, PdCNH^jjClg, is analogous to the classical Werner complexes of platinum. It is of little interest in the catalytic chemistry of 17 palladium. Many other bls(alkylamine)palladium chlorides are known, e.g., Pd(RNH2>2 , R=NH3, CH3NH2 , CH3CH2NH2 and PhNHg.55 In every case, the trans isomer is more stable than the cis isomer, which is usually difficult to synthesize. A thorough study on a series of PdL2Cl2 complexes has been done by far IR spectra^ recorded from 600-250 cm"'' . The position of the Pd-Cl bands are correlated with those expected from a trans effect which is due to both bond weakening and 7T-bonding phenomena, depending upon the type of ligand. Ligands may be arranged in order of their trans directing strength for palladium(II) complexes as follows:

NH3 < RNH2 < pyridine < Cl" < Br" < I" “ RgS

On the other hand, cis isomers are formed readily with bidentate ligands such as the following structures show C4-8).

Cl .Cl o2n / pd; Pd Cl 3 0 '2 ■T l2 ^ Cl 5 /V Cl 18

2.3 Applications of organopalladium complex In homogeneous catalysis: The organometallic chemistry of transition metals has experienced one of its most exciting advances during the past decade, i.e., the conversion of homogeneous reactions requiring stoichiometric ratios of organometallics to catalytic reactions, largely decreasing the amount of noble metal needed. According to T.H. Black's generalization,57 1 homogeneous catalysts possess several important advantages over their insolubilized counterparts— heterogeneous catalysts. These can be summarized as follows: (1) Each expensive metal atom is an "active site", as opposed to just those on a surface. (2) Each atom is in an identical environment, Increasing reaction specificity. (3) Selectivity can be "fine-tuned" by the judicious choice of ligands, solvents, and other variables, resulting in pure products SCHEME I

Alkoxycarbonylation Carboxylation r c o 2h Carbanion RC0NR2 + M-H <" RC-M- Nu: r c o 2r

Decarbonylation CO Insertion

! F Reductive Elimination

R M

Oxidative R-M-X RX + Addition I R=Aryl,(Arylatlon) 4 Vinyl,(Vinylation Alkyl,(Alkylation Elimination

R'M’ M'X + MRR' Trasmetallation M '=MgSe,HgX,SnX.

\—/ . n-H Coupling Reaction R R ’ -> R-R 20

in high yields. (4) Heat is more efficiently dissipated, and the reaction conditions are generally milder and lower energy processes. (5) Mechanistic studies are easier, allowing better understanding and thus better control of reactants. Our understanding of mechanism is quite in the "elementary school" stage. However, the development of transition metal-mediated organic synthesis has enjoyed explosive growth. Pd, one of the most versatile group VIII metals, has been successfully applied to the synthesis of a variety of organic compounds. Basically, both fZ - and -organometallic complexes are involved in catalysis, y -complexes usually arise from the oxidative addition of a PdCO) complex or metal to organic halides. 7 -complexes undergo several major reactions, as summarized in Scheme I. -complexes, are subjected to nucleophilic attack by WgO, amines, etc. (Scheme II.) Typical palladium-catalyzed reactions to those related to this research are surveyed further. 21

SCHEME II

+ PdCl- "* Pd c ( \ i

R- NH CH3 CHO + Cl-Pd-K

nr2 = /

Nu: -Pd Pd -OAC Nu'

SCHEME III

Precursors of Arylpalladium Intermediates

ArTeX ArHgCl

c ArPdX ArH

ArTlX.

ArSiX. RMgX ArSnX: ArSiX ArSnX 22

2.3.1 Arylation of alkenes: Scheme III summarizes the sources of arylpalladium intermediates which arylate alkenes. Reaction a,d,e,f,i,j and k represent transformations which require a Pd(II) catalyst. Reaction c gives a direct palladation of an aromatic compound from a PdClI) species. However, reactions b Caryl halides) g, and h (aryl pseudohalides) require a PdCO) catalyst. Reactions a, b, and c are the most common methods to generate arylpalladium intermediates. The detailed mechanism of palladium-catalyzed vinylation of aryl compounds is not known, but a fairly accurate approximation can be made, based on products obtained and reagents added, also the large amount of information available from studies of other organopalladium reactions. Heck suggested the mechanism eg shown in Scheme IV. Thus, palladium-catalyzed vinylation of aryl halides can be represented in a catalytic cycle as shown in Figure 1. SCHEME IV

O.A. a)ArX + Pd(0)Lo £ ± ArPdL2X

* c h 2=c h r ’ -L b)ArPdLpX + R'CH-CH2 £ Ar^Pd-X

syn-addition

H R H R’ \ _ / Syn- Elimination Ar—HU---C— H / C I \ " \ " Ar 1 H H PdLX H-Pd-X L /

+L H R' - \ = / + -H-Pd-X A/ \ L L L,Base c)H-Pd-X j ~ i PdLn + BaseH+X” L FIGURE I

Palladium-catalyzed vinylation of aryl halides

Oxidative Addition

H-Pd-X ArPdL_X

R'CH=CH,

Ar-Pd-X syn-Eliminatio; Syn-addition — C^H lt TdLxl 25

Either Pd(II) complex or salt or palladium(O) complex can be introduced to start the catalytic cycle. Most often, Pd(II) complexes or salts are used and reduced to Pd(0) presumably by oxidizing some of the olefin present. The palladium(O) complex in solution undergoes oxidative addition to aryl halides to afford the organopalladium intermediates. Upon complexation with olefin, the resulting organopalladium adduct undergoes aryl group migration (it , cr interconversion) to give the corresponding cr -complexes. The CT -complex is believed to undergo an elimination process to generate a TU-complexed hydridopalladlum halide or association with another ligand, the final product, stilbene, is extruded from palladium sphere. In order to recycle the catalysis, a base is necessary to regenerate the PdCO) species by neutralizing the HX bound to the catalytic site. Once PdCO) complex is regenerated, it will react with aryl halides and the cycle begins again. The relative reactivity of aryl halides being oxidatively-added to (Ph^Pj^Pd is in the following sequence: Arl > ArBr >> ArCl. Fitton and RickJJ suggested that this may be another example of aromatic nucleophillc substitution in which the bond-breaking to the leaving group is the rate determining step. 26

(Equation 7.)

Pd(FPh3)4 +

r R

The regiochemistry of phenylation of alkenes is consistant with the fact that steric factors predominate (Table I). At the same time it also reveals the existence of an electronic effect superimposed upon the normally dominant steric effect. The percentages at the different carbons indicate the phenylation distribution.

TABLE I Regiochemistry of Phenylation

ch2 =chco2 ch3 ^h2=chcn ^h2 =chc6 h5 100 100 100

CH2 -CHCH(OCH3 )2 CH2 -C(CH )CfiH5 100 100 3

CH2*=CHCH(0H)CH. T T 100 90 10

60 27

The direction of addition of organopalladium in intermediates to the olefins is generally predictable on the basis of steric effects, e.g., the organic group prefers the less substituted carbon of the double bond.

Electronic effects are also quite significant in some cases. Electron-donating groups (EDG) on the double bond tend to induce more addition to the most electron-deficient double bond carbon (10). The presence of an electron withdrawing group (EWG) on the alkyl side chain of double bond enhances the selectivity (9). The electronic effect correlates with the expected polarization of the Ar-Pd bond, i.e., the aryl group is added to the most electron deficient and the palladium to the most electron rich carbon. The stereochemistry can be readily explained by a syn addition of organopalladium compound followed by a 28 syn elimination of the palladium hydride. For example, bromobenzene and (Z)-1-phenyl-1-propene produces 73? ( )-1,2-diphenyl-1-propene** (11) (equation 8), while the E olefin gave 79? of the (E)-1,2-diphenyl-1-propene (12).11 (Equation 9.)

Pd(OAc)

(9)

H5c6 12

Generally, lower reaction temperatures enhance the stereo- and regio-specificity. Utilization of bulky palladium ligands, i.e. triphenylphosphine, has the same effect. A wide variety of substituents may be present on the aryl halides used for coupling with olefins. The reaction occurs in the presence of chloro, cyano, carbomethoxy, carboxy, aldehyde, nitro, dimethylamino, hydroxy, amino, methylthio, acetoxy, and polynuclear aromatic groups. The olefins employed were limited to styrene, 4-nitrostyrene, 1-phenyl-1-propene, ethylene, propylene, 29

1-hexene and methyl methacrylate. Yields were reported to be low for olefins having methylene groups adjacent to the double bond. The order of alkene activity has been established in the reaction of phenylmercury(II) chloride catalyzed by LigPdCl^ with olefins^ as given below.

CH2=CH2 > Me02CH=CH2 > MeCH=CH2 > PhCH=CH2 > PhMeCH=CH2

14,000 970 220 42 1

Substitution of alkenes either with electron donating or electron withdrawing groups slow down the arylation rate. Steric hindrance has a dominant effect where ethylene is the most active alkene. A variety of styrene derivatives were prepared by c palladium-tri-o-tolylphosphine catalyzed reaction of ethylene and aryl bromides.^0 For example, a study of the synthesis of 2-methyl styrene from 2-bromotoluene and ethylene, as shown in equation 10, showed that five-six atmospheres of ethylene pressure were necessary to obtain good yields. 30

Pd(OAc)2 (PAr3)

The yields of 2-methylstyrene (13) in 20 hr. increased from 54 to 83 to 86? as the pressure of ethylene was increased from 20 to 100 to 120 psi respectively. At lower pressures, E-2,2,-dimethylstilbene (14) was formed as a side product by a second arylation of 2-methylstyrene. However, at pressures above 200 psi the reaction rate decreased. At 750 psi 60? of the starting bromide remained unreacted. The catalyst is believed deactivated by coordination of ethylene. In addition to diarylation, some side reactions stem from polymerization. In some other cases, o-divinylbenzene 6 0 was obtained from o-dibromobenzene in 76? yield. Styrene derivatives with methyl, nitro, acetamlde, amino, formyl and carboxyl substituents were prepared PhgP is the most commonly used ligand for most palladium-catalyzed vinylic substitution. However, when aryl halides with strongly-donating, substituents are 31

p £ <1 treated with methyl acrylate, 1 the reaction mixture is initially homogeneous, but soon begins to deposit black palladium metal. The reaction ceases when all the soluble catalyst decomposes. It was found that reductive elimination of aryl and triphenylphosphine from the arylpalladium intermediate results in the formation of a phosphonium salt (15) as shown in equation 11.

PPh-j Ar.j,d-x --- - 2 ------> ArPPh3+X- - Pd(PPhj)2 (11)

PPh3 12

Quaternization leaves weakly solvated palladium atoms which collapse quickly into inactive metal. From the data in Table II, it is apparent that electron-donating substitutents enhance the quaternization process, whereas electron-withdrawing substituents retard it. From the study by Heck on the effect of the triarylphosphine on vinylic substitution with 4-bromophenol, it was suggested that one possible solution to Pd catalyst deactivation is to use a phosphine with electron-withdrawing substituents. Attempts were made with either tris(3-trifluoro- methylphenyl) or tris(4-carbomethoxyphenyl)phosphine. 32

Table II

The relative rates of phosphonium salt formation (100°C). relative rate aryl halides triaryl phosphine solvent based on

M-HOC6H4Br PPhg CH^CN 3.2

M-H2NC6H4Br PPhg CH^CN 3.2

4-CH3C6H4Br PPh3 CH^N 2.0

C6H5Br PPh3 CHgCN 1.0

3-CF3C6H4Br PPh3 CHgCN 0.2

2-Bromopyrid ine PPh3 CHgCN <0.1

(CH3)2C=CHBr PPhg CH^N 0.5

CgHgBr P(o-tol)3 DMAA <0.1

4-H0C6H4Br P(o-tol)3 DMAA <0.1

C6H5Br P(i1-(CH3)2NC6H4)3 DMAA 6.4 33

Both cases proved unsuccessful. Another possible solution was to inhibit quaternization by increasing the steric hindrance around phosphorous. 2-methyl, 2-ethyl and 2,5-i-pr2 aryl phosphines were investigated. The tri-o-tolylphosphine apparently inhibited sterically the reductive elimination. It also appeared to accelerate the overall reactions because the large o-tolylphosphine promoted dissociation and thus enhanced the coordination of the olefin. However the palladium catalyst remained in solution no evidence for degradation was observed. The larger ortho substituants, ethyl or isopropyl were less effective because the palladium intermediates were less stable than o-tolyl complex. (Table III). Table III Influence of substituted triarylphosphine ligands on vinylic substitution of methyl acrylate with 4-bromophenol.

PR 3 P;Pd Reaction Time Temp % Yield (glc)

PPh3 2 48 hr 75°C 3 6 100 hr 5 P(2-CH3C6H4)3 2 38 hr 75°C >95 6 48 hr >95 PCS-CpH-CfiHn)- 6 50 hr 75°C >95

P(2,5-i-Pr2C6H3)3 6 50 hr 75°C 68 34

The use of P:Pd in a 6:1 ratio tended to increase the overall yield by keeping the Pd complex soluble and stable, but it slowed the reaction rate to a certain extent. There are some cases where aryl halides react poorly or fail with methyl acrylate as shown in Table IV. Table IV Aryl Halides exhibiting limited reactivity Aryl Halides Product Observation

C02H Probably forms Pd chelate None methyl ester gives 69?

Hr oh Acetate gives 69? 24? Iodides without PR^ gives 95? &

s N-acetyl M-rfl derivatives gives J 48? 83? □ (Po^tol3/Pd=4) NH.

Br

None Diacetate gives 26? HO

Br

None Diacetate gives 2?

OH 35

The carboxylic acid group in o~bromobenzoic acid is in close proximity to the Pd center, a very stable inert chelate is formed. Converting the acid into methyl ester reduces the stability of the Pd chelate and the expected product is isolated. o-Bromophenol and activated bromides give only 243E and HB% yields, respectively, possibly because of quaternization of the arylpalladium-phosphine intermediate. Apparently the competing quaternization of the phosphine deactivates the catalyst. Acetylation of p-bromoaniline substantially improves the yield of the desired product. Another synthetically interesting application is the arylation of allylic alcohols to afford fi ? phenyl-substituted aldehydes and ketones, as shown in equation 12.

R Pd 0

The reactivity and regioselectivity depends on the structure of allylic alcohols. The formation of aldehyde or ketone requires several intermediate steps, e.g., an initial elimination and a readdition of the hydride in the reverse direction, followed by another elimination of metal hydride toward the hydroxyl-bearing carbon as shown in Scheme V. 36

SCHEME V

Mechanism of Arylation of 5-buten-2-ol

r 'o h PdX2 (PR3 )2 * Pd(FR,) n=2,3 or 4 EtjN.PR^ 5 n Pd(pR3)n+ ArX - -»Pd{Ar)(X)(PR3 )2 + (n-2)PR3

Pd(Ar)(X)(PR3 )2 + CH2 =CHCH(OH)CH3 * [(R3P)2 (X)PdCH2CH(Ar)CH(OH)CH33 I [ ArCH2^HCH(OH)CH33 I I Pd(x)(PR3)2 -PR At c h 2=c (a t )c h (o h )c h 3 c h 3c -c h (o h )c h 3

H-Pd-X X-Pd-PR, pr3

-[HPd(X)PRj •[HPd(X)PRJ I CH_=C(Ar)-CH(OH)CH, CH,CH(Ar)-8cH,O D < -PR, II ArCH|CHCH(OH)CH3 ArCHCHgCH(OH)CH3 H-Pd-PR, X-Pd-PR, X 5 L III -PR, ■[HPd(X)PR3]

ArCH=CHCH(OH)CH,

ArCH2 CH|C(0H)CH3 -[HPd(X)(PRg)) X-Pd-PR -» ArCH2CH28cH3 H 3 IV 37

The addition of a methyl group at C-3 of allyl alcohol considerably slowed the rate of reaction and increased the yield of 2-arylation. The addition of a methyl group to C-2 of allyl alcohol produced a more significant change. 3-phenylpropionaldehyde was the dominant product; only minor amounts of 2-phenylpropionaldehyde were detected when p-iodoanisole was treated with 3-buten-2-ol, the p-methoxy group had very little effect on the isomeric distribution of * products, and the ratio of 3- to 2-arylketone was slightly lower than that obtained from iodobenzene. The reaction of p-trifluoromethyl bromobenzene with 3-buten-2-ol was also investigated, in contrast to the' p-methoxy electron donating group, the p-trifluoromethyl group gave a slightly higher ratio of 3- to 2-arylketone. Intramolecular cyclizations have also been studied in synthesis of heterocycles, such as indoles (equation 13» 11! and 153*64,65 qUinoiines (equation 1 6 ) , ^

isoquinolines (equation 17) and maerocyclic lactones (equation 18).^ Except for maerocyclic lactones, indoles, quinolines and isoquinolines are cyclized products derived from derivatives of o-haloarylamines which are common reagents. 38

Pd(o) Cat. Ac Ac

I

-COpMe

+ Pd(0) AC o 6

(13) Ac

Br Pd(OAc)2 (1*0

PPhj , 110 *c H Y H

Pd(OAc)? £— > (15) PPh,,NaHCO, R" 5 2 R" R" R* R

1 02Et Pd(0Ac)2 COgEt (16) OCX PPh^.llO'c H

Ph

)2 Pd(OAc (17) PPhj.TMEDA '"sP h v ' P h 39

In an attempted synthesis of indoles, N-(2-cyclohexyl)-N-2-bromoaniline (16) and N-(2-methallyl)-N-2-bromoaniline (17) fail to cyclize. For the former case, the reason for the lack of reactivity is not apparent; in the latter case, closure to a five-membered ring would produce a CT-alkylpalladium complex lacking -hydrogens and thus lack of the ability to regenerate Pd(0) catalyst. By introducing an electron-withdrawing group on the allyl substituent, 2-bromo-N-(2-carbethoxyallyl)aniline (18) cyclizes successfully to 3-carbethoxy quinoline (19), as shown in equation 19.

0

PdCX? (WeCN) (16)

IS 17 C O E t

16 19 40 SCHEME VI Precursors of Vinylation Intermediates

Y

Y

a Br

Y X Y c A1R2 g PdX

Y

'SiRA-

‘3

2.3.2 Vinylation of alkenes: Palladium-catalyzed vinylation of alkenes is a reasonable extension of arylation of alkenes. The necessary vinylpalladium complexes, a direct extrapolation of the arylpalladium complexes, are formed fiT fifi as shown in Scheme VI. Among them, path a, ' b, 6q and c are the most common approaches. Vinylic bromides react analogously to aryl bromides with @ -unsaturated esters,2 »^i70-72 styrene and

ethylene*^0 in the presence of tertiary amines to produce dienes. For example, 2-methyl-1-bromo-propene reacts in palladium-catalyzed vinylic substitution with 2 methyl acrylate to afford the corresponding E-diene under the standard reaction conditions for arylation, however, the stereospecificity of reaction with (E)- and 41

(Z)-l-hexene iodides and methyl methacrylate170 is low compared to arylation. Under the same condition, however, olefins not containing the strongly electron-withdrawing carboxyl group often fail to react. The reaction is envisaged to follow the mechanism in Scheme VII. The initial syn-addition of vinylpalladium complex (20) to alkene and the palladium hydride elimination occur to give complex (21); readdition of palladium hydride in a reverse sense to produce a palladium e3~- allylic complex which cyclizes into a 1C -allylic palladium intermediate (23) can be the source of an (E,E) diene (24). If complex 21 reacts with excess phosphine before readdition and isomerization occurs, the (E,Z) diene isomer (22) is formed. The use of secondary amines, such as piperidine and morpholine, as bases to effect the elimination reaction controls the product distribution. The stronger nucleophilic secondary amines decompose the very stable -allylic palladium complexes. For example, cis-1-bromo-1-propene (25) and acrolein dimethyl acetal react in the presence of piperidine to give only 71 terminal addition products, as shown in equation 20. i\ 2

SCHEME VII c=c- COgCH^ * "“" V * — * S = \ -PPh3

20

CO-CH H-Pd-Br PPh,

o2ch3

23 (E,Z) 22 II PPh3 H-Pd-Br / C02Me

(E,E) 2U

H 1# Pd(0Ac)2 F=\ r ^Br + CHfOCH^), ,0 2% P(o-tol)j (20j

,CH(OCH )2 !k*Br~ o 27 43

The diene obtained is exclusively the E,Z isomer (26). Tertiary amine (27) is the major by-product generated as the result of piperdine nucleophilic attack on the cr -allylic palladium intermediate. Table V shows the ratios of terminal to internal addition to 1-hexene under similar reaction conditions for arylation.

Table V Organic Bromide Ratio of terminal to Internal addition to 1-hexene

ir >20

>20

18 itn

17 ) ■ %Br

® 3 ° 2 C H6 ^ B r

Vinyl bromide and o( -methyl substituted vinylbromide exhibit high regioselectivity, P -monosubstituted and disubstituted vinylbromides yield only a minor excess of terminal addition products. One interesting application of this reaction is the synthesis of 9,1l-dodecadien-1-ol, a pheromone of the red ballworm moth (Scheme VIII). 45

SCHEME VIII

Br o H

CH3I NaOH i d ] o (CH2 )qOH 29

2 )70H + (c h 2 )6o h

30 31

A mixture of readily available 9-decenol (28) and vinyl bromide is allowed to react in the presence of a palladium catalyst and morpholine to yield

l2-morphollno-1O-dodecen-l-ol (29) which was then subjected to a Hoffman degradation to give a 70:30 mixture of (E)- (30) and (Z)-dienol (3D in fair yield.73 According to Patel and Heck, 74 the following trends will be valuable in predicting the products which will be formed in a variety of vinylic halide-olefin reactions. (D Vinylbromide and oi -substituted vinylic bromides add exclusively to the less 46

substituted carbon of olefins. (2) /^-substituted vinylic halides tend to add preferably to the less substituted carbon of olefin. Selective attack at the carbon bearing electron withdrawing group (e.g. ester) will occur if the site is not blocked by steric hindrance. (3) Unhindered secondary amines generally retain the stereochemistry of vinylic bromide substituted into dienes. (4) An amine adduct is usually formed by attack of the amine at the least substituted or the least hindered end of the Tc -allylic intermediate. The successful use of morpholine and piperidine in palladium catalyzed vinylic halide-olefin reaction was thus extended into vinylic halide-diene systems. Vinyl bromide, E-1,3-pentadiene, and morpholine reacted in the presence of a 1 mol? palladium acetate-2 mol? tri-o-tolylphosphine catalyst in 17 hr at 100°C to form 6-morpholino-1,4-heptadiene in 18? yield.The remainder of the product was polymer and higher boiling materials. In contrast to the formation of a it -allylic complex of Pd by a palladium hydride elimination-reverse readdition two step process, for the conjugated diene the tl -allylic complex (34) is formed directly from a

vinylic addition and the subsequent complexation of

double bonds with Pd. Scheme IX

‘Pd(PR, )-,Br

R,P Br

Conjugated trienes (33) were often obtained as minor products in the palladium-catalyzed reactions of vinylic bromides with conjugated dienes and secondary bases (piperidine or morpholine). The it -allylic intermediates underwent nucleophilic substitution with the secondary amine present to produce amino adducts more readily than elimination to afford the trienes. Introduction of a terminal carboxyl group in the diene (35) promoted the formation of conjugated triene (36) in reasonably good 48

yield when triethylamine was used as the base, as shown in equation 21.'75

PdCOAc);, (2 1 ) P(o-tol)

The presence of an aromatic ring conjugated with the double bond in vinylic bromide, presumably, contributes to destabilize -allylic complexes. The anion conjugated with either phenyl or carboxyl group will be formed readily by proton abstraction by triethylamine. The highly conjugated systems prefer elimination to substitution.

A few syntheses of conjugated tetraenes have been attempted.The reaction of (E,E,E)-methyl 2-methyl-2,4,6-octatrienoate (38) with (E)-methyl 3-bromo-2-methyl-propenoate (37), an active vinylic bromide, afforded a tetraene (39) in 28% yield as shown in equation 22. 49

EtjN, Pd(OAc)2

P(o-tol)3

.C0£H3

(22) 39

Based on the study of UV and NMR, the product tetraene

(39) is believed to be the E,EfE,E, isomer. The cause of the low yield is due to competing polymerization reactions. However, the tetraenes studied in most cases possess high melting point and are nicely crystalline compounds which are easily separated from the rest polymeric mass.

Equation 23 shows one case of the divinylation of 1,3,5-hexatriene (41) with (E)-methyl 3-bromo-2-methyl propenoate (40) to give a pentane (42). The all-trans-pentaene diester, orange-red needles, was isolated in 36? yield. The structure was assigned by the study of UV, NMR annd MS.

Pd(OAc)2 2 + ■* p(o-tol) 41 3

2 ^ 3 + 2 Et,,NH+Br (23)

2.3*3 Palladium-catalyzed nucelophilic additions to 50

alkenes: One event that stimulated the development of palladium chemistry was the discovery of the Wacker process in 1956 by a German research team^ (equation 24).

PdClp CHp=CH, CH CHO (24) * 2 2 CuCl2 5

The water is the nucleophile that attacks an intermediate of (it -olefin)palladium complex. These sequential oxidation and reduction constitute a catalytic cycle as follows. (Scheme X.)

SCHEME X CH2=CH2 + H20 + PdCl2 ----- *CH3~CH *CH2+ Pd(0) + 2HC1

Pd(O) + 2Cu C12------^PdCl2 + 2CuCl 2CuCl + 2HC1 + i O p ------> 2CuCl. + H-0 ^ 2 2

h2c=ch2 + i o2 ------» ch3ch=o

In Wacker process, ethylene is oxidized to acetal- dehyde which is a very important industrial process. Alcohols^'^® and acetic acid*^'^® react with ethylene similarly to produce enol ether and vinyl acetate respectively. With larger chain mono-olefins, under similar conditions to Wacker process, water attacks exclusively at the more substituted position and ketones produced. This method has been applied to the synthesis of polycyclic materials (equation 25). 7Q

When olefin and hydroxyl are present in the same molecule, nucelophilic attack of ( ft -olefin)palladium complex by ortho-OH will result in an intramolecularly-cyclized product with 5- or 6- membered rings.2-Allylphenol (43) itself cyclizes to Op 2-methylbenzofuran (44) in equation 26.

Pd \*i Pd(0Ac)2 1 Cu (0Ac )2 , o cOH " 43 (26) -PdH O q l 44 When Pd(II)salt contained a chiral ligand, the optically active cyclized product (45) was obtained (equation 27)

(+)-(fl^-plnene)palladium- l2 (27) acetate, 02> CufOAc)^ H

Intramolecular cyclizations in which the nucleophiles is a carboxylate ion produce unsaturated lactones (46) (equation 28).

Nitrogen nucleophiles normally complex much more efficiently to Pd(II) complexed than oxygen nucleophiles which are bound loosely to Pd(II) complexes. Amines and most other nitrogen nucleophiles are excellent ligands for palladium(II) complexes and readily displace olefins from the metal.

There are several interesting features for amination of olefin, (Scheme XI). 53

SCHEME XI

Cl

* Pd

49 R 2 R..NH ,Pd -50'c Cl n h r ;

Pb(OAc) CO

\ J C 1 Br2 , NBOH -'S k 'OMe (Ac The reaction must be carried out at low temperature (-50°C) with careful addition of amine to the preformed olefin palladium(II) dimer (47)* At temperatures above -50°C, an immediate displacement of olefin from Pd(II) complex occurs to produce "inactive” (RgNfOgPdClg species (50). The second amine attacks olefin in trans stereochemistry, and HC1 generated will be trapped by the third amine. It is necessary to have three equivalents of amine to ensure a high concentration of the relatively unstable aminoalkylpalladium complex (49). This compound could 86 be isolated and characterized by NMR, and could be further converted into several other stable aminated products, 51, 52,87 5387 and 54.88(89 In contrast to the difficulties encountered with the intermolecular amlnation of olefins, intramolecular amination(cyclization process) are much easier, especially because aromatic amines are less basic than aliphatic amines by a factor of 108 . Indole (56) forms readily from 2-allylaniline (55) (equation

29).90 55

PdCl_(MeCN)2r--- 7 . Quinoline C X x ‘29> H 55

In equation 30, the more basic primary cyclopentyl amine-olefin system forms a very stable olefin-amine-palladium(II) complex (57). Because the amine nitrogen is strongly coordinated to metal, it can not attack the coordinated olefin to complete the cyclization step.

(II) 57

The problem can be overcome by conversion of the free amine to its tosamide (58), a much weaker base (equation 31). The cyclized product (59) was reduced by photolytic detosylation to afford the free cyclic Q1 amine (60). A number of other olefinic amines and amides which cyclize in the presence of palladium(II) complexes were reported 56

Pd(II) 1)H2 /Pd, 2 )light ROH H Ts

59y*-* U s , J U <51> H 60 y«rf

When the nucleophile is a carbanion, another difficulty occurs; Pd(II) is reasonably strong oxidizing agent for the oxidation of carbanions. In the reaction of methyllithium with styrene in the presence of palladium(II) salt, the most oxidation-resistant complex, palladiumCII)acetylacetonate, gives the best yield.97

However the use of the stabilized carbanions (Pkaa = 10-18) as nucleophiles causes a change in mechanism. The nucleophiles attack externally at C It -olefin) palladium(II) complex (61) at the more substituted position of the olefin. (Scheme XII) 57

SCHEME XIV R X X - * ^ R THF + PdCl2 (CH3CN)2 + Et3N + J r Y -60 c

X R

T — par- I 61 .

R R

d-

R = H, Me, Et, n-Bu, NHAc R' = H, Me, n-hex X = COgMe, C02Et, COMe, C02-t-Bu Y = C02Et, C02Me, COMe, Ph

The addition of two equivalents of triethylamine is required to afford an active olefin palladium(II) amine complex prior to addition of the carbanion. Terminal 58 mono-olefins are alkylated in almost quantitative yield with alkylation at the most substituted carbon. Electron-rich olefins give high yield, but electron-poor Q8 olefins such as acrylates do not alkylate at all. The regiochemistry of this alkylation is consistent with the mechanism proposed by Heck for vinylic substitution of aryl halide. Palladium-catalyzed vinylic substitution of aryl halides and vinyl halides in which "carbanion” nucleophiles are generated by oxidative addition process have been discussed earlier. The nucleophilic addition of a variety of nucleophiles to ( ft -olefin)palladium complexes lead to the creation of new carbon-carbon bonds, carbon-nitrogen bonds and carbon-oxygen bonds. Trost summarized various nucleophilic additions to olefins catalyzed by palladium in a diagram as shown in

Scheme XIII." An important question in nucleophilic addition to a coordinated alkene concerns the regio- and stereo-chemistry of the addition. As can be seen from Scheme XIV, the nucleophiles can attack the coordinated alkenes at four different sites and hence four different isomers may be obtained. 59

SCHEME XIII Additions to Olefins catalyzed by Palladium CH_CO H CH,CHO

OHCCHO

CH-=CH -> CH CH=CHCH_ ClgCHCHO 2 2 3 3 CH_CH„CH=CH,

c i c h 2c h o CH-CH(OR)

NuA SCHEME XIV 2 1 R 1 / R2 HuA d-NuA R NuA u a trans /R2 ** ** attack 2 Pd-NUA H

HuB

pd-NuB

NuB 60

According to the studies of the’stereochemistry of the addition of a variety of nucleophiles to ( it -olefin)palladium complexes, one can establish two classes of nucleophiles; nucleophiles that add in an external trans attack (Nu A, Scheme XIV), and nucleophiles that add through internal cis migration from the metal to the carbon (Nu B, Scheme XIV). Nucleophiles that add trans(Nu A) are water (hydroxide),100*101 carboxylates,102 alkoxides,1 °^*10i* amines,10^ and stabilized carbon 1 nfi nucleophiles; nucleophiles that add cis (Nu B) are hydride,10^ aryl,10^ ’10® and alkyl.10^ Despite the fact that oxygen and nitrogen nucleophiles are coordinated to the metal in (Tt-olefin)palladium complexes, they do not undergo a migration from metal to the coordinated olefin. According to the theory by 110 Fukui, one may divide the system into two parts, HOMO and L0M0 as shown in Figure II.

FIGURE IZ

u m o (*")

HOMO Pd

111 112 In a perturbatlonal treatment, Klopman, * 61 has classified reactions as frontier-controlled and charge-controlled. When the energy difference

^"^LUMO-HOMO *5 ^ar6e » reaction becomes charge-controlled (trans addition, Scheme XIV), but when the energy gap is small, the reaction becomes orbital-controlled Ccis migration). Based on the above statement, palladium-catalyzed vinylic substitution of aryl halides or vinyl halides is orbital-controlled cis migration addition to ( 1L -olefin)palladium. III. Polymer Supported Catalysts 3.1 Introduction Customarily catalysts were classified as homogeneous (soluble) and heterogeneous (Insoluble). The use of both catalysts is almost as old as chemistry itself. Acid catalysis is a typical example of homogeneous catalysis which has been used for many years, on the other hand, the hydrogenation of unsaturated compounds by metals such as platinum, palladium or Raney nickel represents the application of heterogeneous catalyst. With the development of deposition techniques, heterogeneous catalysts have been modified by the absorption of dispersed metallic atoms over cheap, high-surface area inorganic supports such as silica, magneta, active carbon, alumina, and zeolites to 113 produce highly activated catalysts. Homogeneous catalysts are usually more active and more selective than heterogeneous ones. All catalytic metal atoms in homogeneous catalysts are equivalent in terms of steric environments and electronic effects. The major disadvantage of a homogeneous catalyst is the requirement for an additional separation step, i.e., distillation, extraction and filtration in order to

62 63

recover the catalyst that makes industrial applications less attractive. The major advantage of a heterogeneous catalyst is its extreme ease of recovery. However a troublesome diffusional limitation imposed by the rigid matrix makes the access of the substrates to nonequivalent active catalytic sites more difficult. A comparison between homogeneous and heterogeneous 1111 catalysts is made in Table VI. Chemical industry uses a wide variety of catalysts which are mainly heterogeneous as shown in Table VII.1^ Prior to the development of modern surface analysis techniques for hetereogeneous catalysts, it was easier to study a homogeneous catalysis in terms of characterization of the catalytic species and the elucidation of mechanism. Thus, the studies on homogeneous catalysts are more advanced in the laboratory research. Then a hybrid catalyst was developed; homogeneous catalysts immobilized on insoluble supports. The "supported homogeneous catalysts" were designed to combine the high selectivity traditionally associated with homogeneous catalysts with the ease of separation Table VI

Comparison of Homogeneous and Heterogeneous Catalysts

Catalyst Properties Homogeneous Heterogeneous Active Centers All Active Atoms Surface Atoms Only Concentration (local) Low High Diffusion Problems Not Present Present Structure Known Not Known Stoichiometry Known Not Known Modification Possibility High Snail Reaction Conditions Mild Severe Catalyst Separation Costly Easy Application Limited Wide Mechanistic Study Easier Difficult 65

Table VII Catalyst Types Employed By Chemical Industry

Selected Processes Products Major Catalysts Used POLYMERIZATION------Ethylene High-density polyethylene, Trialkylaluminum and Propylene polypropylene titanium(IV) halides (Ziegler-Natta types) Ethylene, vinyl Low-density polyethylene, Organic peroxides chloride polyvinyl chloride, styrene polystyrene

SYNTHESIS GAS- Steam Hydrogen and carbon Nickel on alumina hydrocarbon monoxide with potassium reforming oxide promoter Ammonia Ammonia Ferric, aluminum, and synthesis calcium oxide with potassium oxide promoter Methanol Methanol Combinations of synthesis aluminum, chromium- (III), cupric acid/ or zinc oxides Olefins Aldehydes Cobalt, rhodium, or ruthenium complexes with carbon monoxide, amines, or phosphines Acetic Acid Acetic anhydride Iodine and complexes of rhodium with carbon monoxide & phosphines Table VII (continued) ALKYLATION---- Benzene Cumene Phosphoric acid on silica Benzene Detergent alkylates Aluminum chloride, fluoboric acid, or hydrogen fluoride Benzene . Ethylbenzene Alumina and silica or phosphoric acid on silica HYDROGENATION-- Benzene Cyclohexane Nickel or platinum on alumina with lithium oxide promoter DEHYDROGENATION- Butane Butadiene Chromium(III) and aluminum oxides Butene Butadiene Calcium and nickel phosphates: ferric and aluminum oxides; or ferric and chromium(III) oxides Ethylbenzene Styrene Ferric oxide with chromium(III) and potassium oxide promoters Petroleum Benzene, toluene, and Platinum and rhenium hydrocarbons xylenes on silica and alumina

OXIDATION----- Ammonia Nitric oxide Platinum-rhodium gauze Ethylene Ethylene oxide Silver on alumina with cesium oxide promoter Methanol Formaldehyde Copper or silver, as gauze or on support; or ferric and molybdic oxides Propylene Propylene oxide Molybdenum, vanadium or titanium naphthenates Sulfur Dioxide Sulfur trioxide Vanadium pentoxide 67

Table VII (continued AMMOXIDATION- Methane Hydrogen cyanide Platinum-rhodium, as gauze or on support; or platinum, ruthenium, or aluminum on ceramic support Propylene Acrylonitrile Uranyl antimonate OXYCHLORINATION------Ethylene Ethylene dichloride Cupric chloride PdCl2 needed for Wacker Process support

of reactants and catalysts normally associated with heterogeneous catalysts. Basically, insoluble supports used can be classified as inorganic and organic supports. The comparison between them was shown in Table VIII.

Table VIII

Comparison of Inorganic and Organic Supports

Properties Inorganic Support Organic Support (metal oxide) (polymer) Thermal Stabilitty Good Modest Mechanical Stability Good Modest to Poor Location of Active Sites Surface Surface and Inside Swollen Matrix Dependence on the Liquid Medium Independent Dependent Hydrolytic Stability of Linkage — Good Ease of Modification Difficult Very Easy 68

Polymer supported catalysts (or supporting homogeneous transition metal catalytic complexes on organic polymers) is a new technique developed in late m * £L 1960*s , since then, it has rapidly become an area of intense research activity. The ligand sphere of the metal attached to the polymer is ordinarily unchanged and the attached complex is "dissolved" in solvent and reactants by swelling the polymer surface. The supported homogeneous catalysts are usually insoluble in any solvent, and in that sense, they are indeed hetereogeneous. 3.2 Synthesis of polymer-supported catalysts. Generally, requirements for polymeric support are: good mechanical properties, such as attrition resistance, porosity and sufficient thermal stability, easily accessible sites (for the subsequent complexation of metal) and lack of solubility in the reaction medium. Basically, there are several methods for the preparation of polymer-supported catalysts. They are summarized as follows. Method I). A preformed polymer is prepared and functionalized by reacting covalently with ligand precursor (containing N, P, 0, or S electron-donating atoms), followed by being complexed with an interested catalytic metal. Method II). Homo- or copolymerization of monomer(s) with ligand substituents which are then complexed to a catalytic metal. Method III). Homo- or copolymerization of monomeric organometallic complexes having polymerizable ligands. Method IV). Absorption of organometallic complexes or 117 dispersed metals directly on polymer support. Method V). Immobilization of ionic organometallic complexes or complexes on ion-exchange resin. 1187, 119,120 1 , 7 121,122 7 Method VI). Radical graft copolymerization (e.g. styrene and 4-vinylpyridine) on non-porous silica spheres - maromolecular immobilization - followed by the complexation with catalytic metal. ’124’12^*1 Method VII). Future possibility - Breitenberger1s functionalized polyethylene supports which are soluble at about 110°C but crystallize and precipitate at room temperature or an isotactic polystyrene where the isotactic runs form single crystals (crystalline domain is insoluble-heterogeneous), while the heterotactic amorphous polystyrene loops may be modified functionalized and finally complexed with a catalytic metal. Polystyrene-divinylbenzene (DVB) resins are the most popular support for general purpose of modification 70

(Method I). A more detailed description of this resin is thus necessary. For polystyrene-DVB resins, there are three major classes which have been used: a) Macroreticular macroporous resins: styrene-divinylstyrene resins with divinylbenzene (DVB) contents of 20, 40 and 60$. These highly cross-linked resins possess a very rigid network, high surface area and the greatest amount of transient pores and macropores. Because of their rigidity, usually, the macroreticular resins preserve their surface characteristics during chemical transformations and catalytic reactions. The high crosslink density also prevents rapid diffusion into the domain of particle core under the internal surface where they are wet by the solvent. Usually the surface of polymeric beads is less highly crosslinked or rigid than the core of beads. 127 b) Microreticular or microporous resins. Styrene-divinylbenzene resins with divinylbenzene (DVB) contents of 1$ or 2$ DVB. These resins are swollen in solvents and, ideally, all of their internal volume is accessible by solvents and reagents. Therefore, they may be functionalized both on the surface and internal matrix of polymers which possess high capacity. c) Between macroreticular and microreticular. 71

Styrene-divinylbenzene resins with DVB contents between 20? and 2?. It is believed, for these resins, that there is no clear distinction to tell whether they belong to a) or b). The properties of c) resins will be between a) and b).

3.2.1 Functionalization of Preformed Polymers (Method I): The most often used preformed polymer, Merrifield resin (62), i.e., chloromethylated DVB-crosslinked polystyrene, is prepared by the reaction of cbloromethyl ethyl ether with polystyrene using SnCl^ as catalyst12® (equation 33).

Eton

By replacement of Cl , a variety of carbon-carbon, carbon-heteroatom (C-N, C-P, C-S and so on) and carbon-metal bonds are formed as is shown in Scheme XVI 72

SCHEME XVI 129

Na MnCCOJc,THF

1 )Na "C ,2)MeLi 130

1)NaCN,2)Na0H/Hz0

132 68 -NH 62 C0-.H 133

70 CH

135 CH-I]{CH:CHi;Cr;)_ 7£ 73

Coordinating polymers 63 and 64 were prepared respectively from the reaction of chloromethylated polystyrene 62 with metal carbonyl anions

Na+[(Cp)Mt(CO)n]“ and Na+[Mt(CO)5]~. n -Cyclopentadienyl ligands were introduced into polymer support by the treatment of 62 with Na+Cp“ (Cp = n^-cyclopentadienyl) followed by deprotonation with MeLi to give 65. Cp, a six electron donor, is a suitable ligand for Co, Ti, Rh. Fe, Mn, Zr, Hf and many others as shown in Scheme XVII.

SCHEME XVII C„TiCl 136

Co_(CO)ci A or HV 137 CH Co(CO)

65 LX+ MCI 3 MCI 77

M=Ti,Zr,Hf 74

Titanocene dichloride, Cp2TiCl2 , was attached to polymer by reacting preformed polymer 65 containing one Cp unit, with CpTiCl^ to give 74. Visible far IR spectra and an electron microprobe X-ray fluorescence study showed that the structure was similar to that of monomeric Cp2TiCl2 and the radial distribution of Ti and Cl were uniform through the polymer matrix. Other carbon-carbon bond formation can be achieved by the reaction of 62 with sodium cyanide in dimethyl-sulfoxide, to give 66. The nitrile groups can be further hydrolyzed to 67i or reduced with LiAlH^ to yield -CH2NH2 groups. Phosphines, the most applied ligand for transition metal complexes, may be introduced by reacting the chloromethylated polystyrene 62 with LiPPhg or

KPPh2 to yield 68. Diphenylphosphine-polystyrene resins have been the most widely used supports for attaching organometallic catalytic complexes (Scheme

XVIII).138"1'19 The preparation of a polymer containing phosphorous-bonded transition metal atoms can be performed by reacting the coordinating polymer with a transition metal complex from which a weaker ligand 75

SCHEME XVIII RhCl(PPhj)j (£)-PPh2)3RhCl Z? IrCl{CO)(PPh3)3 ( P y PPh2)2Ir(CO)Cl 79

RhH(CO)(PPh3)3 (P)- PPh2) xRhH (CO) CPPh„ ) y 80

pd(PhCN)2Cl2 (7)-PPh2)2PdCl2 81

RuCl-j.DMA Q - P P h 2)2RuCl3 §1

Rh(CO)2Cl2 /"N X=1|2 [F V-PFh ) Rh(CO)„Cl 83 2 x y y»2(1 RuCl2(CO)2(PPh,i>3 (J)-PPh2)2RuCl2(CO)2 84rite Ni(CO) PPh, 0 - F P t . 3 3 4 (T)-PPh2Ni (CO) 2 (PPh3) 85

Mn(CO)5X ( p ^ - PPh2(cis)Mn(CO)^X 86 Rh(CO)2(acac) X=Cl,Br,I * ^ p V- PPh2Rh (CO) (acac) 87

Co (NO)(CO)3 /—v x=1,2 (PVPPh) Co (NO) (CO) K s x y y=2,1 86 Fe(CO)3 -orkn x=1'2 ( pVPPhJ Fe(CO) 69 t x J y=4,3 Ni(CO)2 (PPh3 )2 ( p ) - PPh2 )2Ni(CC)2 90

3 6 Cr(CO) (1) -arene) 91 ^ ) - PPh2Cr(C0)2 (arene) l_l_r

Mo(CO)& a or hi/ @ - PPh2 )xMo(CO)y 92 76

(e.g. CO) is replaced (85 to 88) or by equilibration with the corresponding homogeneous complex (78 to 84). In some cases, the replacement proceeds photochemically (89 to 92). Due to the constraint of polymer backbone, there is higher tendency to generate multidentate chelated phosphine-metal complexes from cross-linked polystyrene than monomeric situations. For example, phosphinated Merrifield resin 62 was allowed to react with (PR'g^RhCl, when the resin was crosslinked with 1 or 2 per cent of divinylbenzene and contained excess bound phosphine moieties, more than one phosphine were complexed simultaneously to the one metal center due to 150 the mobility of polymer matrix.

@ - ( pR2)n + (PR’3)3PhCl > ( ^ - ( pR2)nRhC1(PR'3)3_n + nPRf3

n = 1-3

When the resin was crosslinked with 20? divinylbenzene and possessed a low phosphorous loading, there was much less multiple site chelation occurring due to the rigidity of polymer backbone and the limited number of phosphorous donors. In addition to Merrifield resin 62, several other types of halogen-containing polymer supports were also 77 employed to attach the phosphorous group, e.g. polyCvinylchloride) , chlorinated polyolefins, poly(chloroprene) , brominated polybutadiene, and poly(p-bromostyrene). Most of the reaction were performed by refluxing the halogen-containing polymers with LiPPhg or KPPh2 in diethylether, THF, dioxane, toluene or decalin solutions. ^Some synthetic problems were found related to the above replacement reaction. Quaternization may occur when Merrifield resins were allowed to react with LiPPh2 or KPPh2 (equation 34) .

'CH. Cl"

CH_Cl

9 1

The unfavorable formation of resulting phosphonium complexes 93 will change the properties of the polymer backbone and polymeric ligand significantly. Direct nucleophilic displacement on brominated polystyrene with diphenyl phosphide was found incomplete with a significant amount of residual bromine. Under more strenuous conditions to obtain complete displacement on poly(vinylchloride), considerable breakdown of polymer icq backbone during phosphination was observed. ? The phosphorus-containing resins, in another approach, can be obtained by the reaction of chlorine-containing phosphorous reagents. PhPCl^ or PhgPCl with polymer backbone bearing nucleophiles, e.g., lithiated polystyrene 95 (equation 35), poly(vinylalcohol) 96 (equation 36), poly(allylalcohol), polystyrene substituted by -CH^OH group as well as natural polymers such as cellulose or starch 160-162

95

PhPci

96

A bidentate chelating phosphine ligand, 97, has been prepared by lithium-bromine exchange (equation 37). The diphos-substituted polystyrene resins were used as supports for rhodium and iron carbonyl complexes.^3-165 79

•Ph PPh2 (37)

97

A chiral phosphinated chelating ligand 99 was also prepared for asymmetric synthesis (equation 38^ 166,167

H 96 99

Nitrogen-containing ligands, just like phosphine-containing ligands, can be obtained by modification of Merrifield resin or other polymer resins with nitrogen-containing precursors. Dimethylaminomethylated styrene-divinylbenzerte was synthesized by the reaction of Merrifield resin with 1 fi ft gaseous dimethylamine. Other polymeric ligands with N-donors were prepared by reaction of Merrifield resins with ammonia, polyamines, potassium phthalimide or sodium cyanide. The synthesis of polymer bonded 1 f\ q i v n porphyrins, s imidazole and phthalocyanines were 80 also reported . Transition metal complexes were introduced to nitrogen-containing polymer by ligand exchange or by a bridge splitting reaction. The polymer is significantly contracted because of an intrapolymer chelation. Numerous N-metal complexes were prepared from poly(vinylpyridine) (PVP), poly(vinylimidazole), poly(ethylenimine) and poly(vinylamine). Dimethylaminated resins, 71, complex with

[RhCCOjgCl]^ by bridge-splitting reactions and with platinum, palladium and rhodium chlorides by coordination. A bidentate polymeric ligand 100 was prepared by a direct nucleophilic addition of lithiated 172 polystyrene to 2 ,2 1-dipyridyl (equation 39).

(39) 100

Poly(styrylbipyridine) (100) has been used to anchor metals such as Cr, Mo, W, Pd(II), Pd, Cr(III), Mn(II), Fe(II), Fe(III), Ru(III), Co(II), Ni(XI) and Cu(II). These complexes proved to be catalysts for a variety of reactions. Oxygenated substituents were also introduced into a 81 preformed polymer chain which may be used for the preparation of polymer-supported catalysts. The oxygen-containing ligands such as ether, hydroxyl, ester, amide and carboxyl were commonly used. Among the natural polymers that can be used directly as polymeric ligands are cellulose derivatives, cotton, wool, silk-fibroin and other polypeptides. Some of commercially available styrene-DVB crosslinked copolymers Cion-exchange resins) bear hydroxyl, sulfonic or carb.oxylic substituents. Most organometallic complexes of polymeric oxygenated ligands were either prepared by homo- or copolymerization of monomeric ligand or followed by complexation with metal ions (method II) or immobilization of ionic organometallic complexes on ion-exchange resins (method V). 3.2.2 Polymerization of Monomeric Ligands (Method II). Method II represents another synthetic approach to construct polymer-bound catalysts involving the preparation of a monomer which contains the ligand, followed by the polymerization of unsaturated monomer to produce the polymeric ligand. Compared to method I, method II has several advantages. 1) A wide range of ligand loadings (concentration) can be controlled by the concentration of monomeric ligand. 2) The nature of polymer, i.e. hydrophilicity and hydrophobicity, can be tuned depending on the nature of the comonomer and by 82 variation of monomeric ratio on copolymeric backbone. 3) The distribution of ligand will be uniform throughout the polymer matrix. 4) Unfavorable side products can be eliminated by purifying the monomeric ligands during their preparations. Styrene, a phenyl-containing monomeric ligand, was homopolymerized or copolymerized with DVB to generate resins. Chromium, molybdenum and tungsten tricarbonyl groups were introduced into polystyrene by forming complexes with benzene rings (equation 40).

K(CO)3

M«Cr,MotV

Vinyl derivatives of phosphines, such as vinyldiphenyl phosphine, allyldiphenylphosphine and p-diphenylphosphine styrene (101) have been radically copolymerized with styrene or DVB to give polymers (equation 41) which were used extensively as catalyst supports. 174,175 83

PPh (41) 0 * 0 * d © - O 102 PPh.c 101

Copolymers of 2,3-di(diethylphosphine) 1,3- butadiene have been prepared and used as chelating polymeric ligand (equation 42).

= r r -> ( CH2-C=j:-CH2 ) (42) PR- PR. R2P PR-

A elaborate synthesis of hydrophilic polymers containing pendant chiral alcohol group and chiral chelating diphosphine group have been achieved (equation 43).

HO >Ph, AIBN

:=0 105 103(5,S)

Me

PPh'

HO Me ■Me

106 84

Both (R,R)-1-methyl-2-hydroxypropyl acrylate (104), which introduces chiral alcohol sites, and an N-acryloyl derivative of a chiral chelating diphosphine ligand (103) were terpolymerized with the cross-linking agent (105), ethylene dimethacrylate to generate resin 106. The rhodium complex of 106 resin was used for 17ft asymmetric hydrogenation. ' 177 Homo- or copolymerization of 4-vinylpyridine, 17 8 N-vinylimidazole ' or acrylonitrile with styrene yields polymeric nitrogen-containing ligands. Poly(4-vinylpyridine), 4-PVP, was most applied to form complexes with a variety of transition metal halides including chlorides of Fe, Co, Ni, Mn, Rh, Pd, Cr, Ru, and Cu. The coordination numbers of the complexes depends upon the central metal ion employed. Substitution reactions of poly(4-vinylpyridine) with cis

[Co(en)2X2 ]X1^^ and trans CCo(en)2X2]X1 produce mono- and di-substituted complexes respectively. Such a result may be attributed to a steric hindrance imposed by polymer backbone, thus the substitution of two Cl- in cis position is not allowed. Poly (6-vinyl-2,2'-bipyridine) (107)^®^ and poly (4-vinyl-4*- methyl-2 ,2 '-bipyridine) (108)1®2 have been synthesized and used to anchor various transition metal salts. 85

108 /V

A polymeric nitrogen ligand was specifically applied as a model for deoxymyoglobin by reacting iron tetraphenylporphyrin (Fe(TPP)) with a polymer-supported imidazole. Other cases containing nitrogen donors are the absorption of palladium dichloride from its slightly acidic solution on poly(cyanoroethylstyrene), poly- (cyanomethylmethacrylate), poly(2-cyanoethylmethacrylate) and poly(allylcyanamide). Interestingly no PdClg complexes were formed with polyacrylonitrile, poly(3-cyanophenylmethacrylate) and polyCfumarodinitrile) in which the cyano group is attached to a polymer chain or a phenyl group; apparently, steric hindrance retards the formation of complexes. Poly(ethylenimine) 1 8U and 185 polyCvinylamine) have been complexed with cupric chloride. Polymerization of O-containing ligands have been performed by radical polymerization and copolymerization with styrene or divinylbenzene with unsaturated 86 functional monomers, such as acrylic acid, methyl methacrylate, itaconic acid, vinyl acetate, vinyl ketone and 1-vinyl-2-pyrrolidinone. Hydroxyl ligands may also be obtained by hydrolysis of the ester group of vinyl acetate and its copolymers, e.g., vinyl acetate-ethylene and vinyl acetate-vinyl chloride copolymers. 3.2.3 Polymerization of Monomeric Organometallic Complexes (Method III) Method III represents advances in polymerization and copolymerization of metal-containing monomers (MCM). According to the types of bonding between ligands and metals, MCM can be classified into three main types. MCM with a covalently bonded (c* -bonded) 109, coordinated (n>'-bonded) 110, and rc bonded metal atom 111. Here M is the metal atom, X the ligand, and Z and Y are functional groups. h2c=ch h2c=ch h2c=ch

! I t Z Y Z

I I ! MXn-1 MXn MXn 109 110 111 M can be either non-transition or transition metals for 109 and 110, or 111? The type of bond has a significant influence both on the stability of MCM and on their nature of polymerization and copolymerization process. Due to the presence of metal atoms, a 87 redistribution of electron density at the growing center during polymerization will have a marked effect on polymerization mechanism. The

(h 9c 4 )3f — » m «— p (ca h 9 )3 X M«Pd(II),Pt(II);X«Cl,Br 112 88

Monomers of the 110 type are polymerized by free-radical i qn initiators in the presence of the metal chloride; * the complex formation and polymerization reactions take place simultaneously. In alcohol solution it is possible to polymerize the following complexes: vinylimidazole - MnCl2 , vinylimidazole-NiCl2 > vinylbenzimidazole-NiCl2 and vinylbenzotriazole-NiCl2 . The homopolymerization of complexes of palladium and nickel halides with vinyl pyridine and phosphorous-containing monomers have also been described. Among the examples of -bonded MCM, 191 vinylferrocene has been most studied. The principal observation from the study of polymerization is that intramolecular electron transfer from the iron atom to the radical center (A1BN as initiator) occurs with the formation of an Fe(III)-containing terminal group, which is then followed by intermolecular chain termination. The radical polymerization of n^-vinylcyclopentadienyltri(carbonyl)manganese (benzene, AIBN)**^ n^-vinylcyclo- pentadienyldicarbonyliridium, and n;6_(benzyl 1 q q acrylate)tricarbonylchromium (THF, AIBN) form polymers with different molecular weights. Most MCM's which are incapable of homopolymerization or polymerization with difficulty, can usually be 89 copolymerized effectively with other monomers. However, c n -vinylcyclo- pentadienyldicarbonylcobalt does not polymerize due to the decomposition of MCM under reaction conditions. Styryltricarbonylchromium does not polymerize by either a free radical or an anionic pathway; the reason is not known. Metal complexes deposited or "exchanged11 onto the polymeric matrixes not by covalent but by Van der Waals force or electrostatic interaction will not be discussed here. (Methods IV and V.) 3.2.4 Macromolecular Immobilization (Method VI) Method VI represents an improvement over the conventional preparation techniques of polymer-supported catalyst. Catalysts on both insoluble polymer supports (cross-linked polymer) and inorganic polymer supports (metal oxide's) have drawbacks such as differences in accessibility, microenvironment, stability, specificity and activity over all catalytic centers. The nonequivalence of active sites hampers a reliable analysis of the kinetic and catalytic processes. To counteract this problem which, at the same time, retain the advantage of easy recovery, non-crosslinked, soluble polymeric catalysts (e.g. macromolecular catalysis) were immobilized on insoluble inorganic supports such as SiOg. The radical initiator attached to silica was synthesized according to Fery, Laible and Hamann 90

(equation 44).^ ^

( s i O ^ - O H )2 + CljSi-CH2 ( Q > ~ N02 ---- »

113

114

The azo initiator, 114, was stored at -20°C and must be used within a few days after synthesis. Styrene and 4-vinylpyridine were copolymerized in bulk with silica-bound radical initiator (114) at 60°C

(equation 45) to generate the grafted silica spheres 115

Monomers 114 + ------60*c

115 91

3.2.5 "Single Crystal" Polymer Support (Method VII) Finally, a future possibility was proposed (Method VII) which makes a polymer readily separable when the polymer can form "single crystals". For example, isotactic polystyrene crystallizes slowly from dilute solutions in benzyl alcohol at different temperatures. The "single crystals" were about 30% crystalline and had thick amorphous layers. 105 The loops of amorphous layers can be modified and functionalized following by complexation of metals.

3.3 Reaction aspects of polymer-supported catalysts. A large number of reactions have been catalyzed by polymer-supported catalysts, and quite a variety of reaction types have also been investigated. Hydrogenation has been the most commonly studied, but 92

other important reactions, involving carbon-carbon bond formation are hydroformylation, hydroesterification, alkene oligomerization, diene cyclooligomerizatlon (dimerization and trimerization are most common), linear oligomerization, alkyne cyclooligomerization, polymerization and alkene metatheses have been achieved. Hydrosilylation, epoxidation, hydration and multistep sequential reactions have also been investigated; some of the reactions have been associated with asymmetric synthesis. In this section, several important properties and characteristics of polymer-supported reactions that can be afforded by polymer supports will be illustrated by specific reactions. The details of each type of catalytic reactions are found in the references cited. Three main features may be controlled by the polymer matrix: 1) selectivity. 2) matrix isolation of catalytic centers, and 3 ) multifunctional catalytic reactions. Each one will be elucidated as follows. 1) Selectivity: This mainly depends on pore size of polymer and steric effect of polymer backbone on the catalytic site. In order to have reaction occur, reagent molecules must diffuse into a swollen polymer matrix to reach catalytic sites. Reagents with different functional groups or molecular sizes will have quite different diffusion rates. By adjusting the pore

_L. size of resins, relative diffusion rates will be changed to a favorable situation where reagents with smaller size will have the highest diffusion rates. The pore size itself can be adjusted by degree of cross-linking, solvent and ratio of ligand to metal. Hydrogenation, catalyzed by polymer-supported Wilkinson's catalyst = ©

-(PPh^RhClCPPhg^i illustrates well the influence of the size of olefinic substrate on the rate of reactlon1^ " 1^® (Table IX). Table IX Relative rates of hydrogenation of various olefinic substrates by homogeneous and anchored Wilkinson's catalyst. Relative Rates

Olefinic Substrates Size of Homogeneous Supported Olefinic Substrates HhCl(PPhg)2 Cataalyst

Cyclohexene Small 1.0 1.0 1-Hexene 1.4 2.55 Cyclooctene 1.0 1/2.54 Cyclododecene 1/1.5 1/4.45 >r ^ --Cholestene Large 1/1.4 1/32

Compared to homogeneous catalyst, anchored catalyst is about 16 times less active but possesses higher selectivity in which the rate of hydrogenation decreases 94 as the size of the olefin increases. This size-rate relationship is, conceptually, the basis for increasing the hydrogenation selectivity on the side-chain double 1 9Q bond over cyclic double bond in a steroid (116).

115

Since the change in the polarity of the solvent causes a change in the degree of swelling, it thus changes the pore size of polymer. The selectivity factors for the hydrogenation of 116 (n=8 , m=0) was enhanced by 3 fold when benzene/EtOH (1:1) was used as solvent instead of benzene. Polystyrene does not swell as well in the presence of polar solvent EtOH as in benzene, so a smaller average pore size results. Polymer-supported palladium dichloride catalysts were employed in ethoxycarboxylation of 1-pentene200 (equation 46). 95

PdCl + CO + EtOH

(n) (b) As the ratio of P/Pd in the resin increased, the normal/branched Cn/b) selectivity increased. This shows the fact that the phosphine groups on the polymer (1%DVB) are mobile enough to allow the bis(polychelated) Pd metal to form in high concentrations which imposes a higher steric effect on the catalytic sites. Olefin-hydroformylation of 1-pentene catalyzed.by polymer-supported RhH(CO)(PPh^)g was investigated by Pittman and coworkers. 201 The catalytic reaction was markedly more selective than its homogeneous analog when the polymer had a high phosphorous loading. Normal/branched (n/b) aldehyde ratios up to 12 were observed when a P:Rh ratio greater than 10 was employed. At a lower P:Rh ratio (3.6), the n:b ratio is about 2 .5-3 for both polymer-supported catalytic and homogeneous catalytic systems. At higher P:Rh ratio, the reaction rate is slowed down by the increase in cross-link density which decreases the polymer swellability and pore size. Copper complexes with polymeric tertiary amines are active catalysts for the oxidative coupling of phenols

(equation 47). 96

Catalyst n '2 0 ---- + n H20 (47)

n

117

Tsuchida and Nishide202 have shown that polymer sterlc effects, which lead to low catalytic activity and low molecular weight of polymers from 2 ,6-dimethylphenol in a poly(styrene-co-4-vinylpyridine)-Cu(II) system, can be released by inserting a spacer between polymer chain and complexing pyridyl groups (118).

2— Cu'

The polymer-supported cupric catalyst (118) provides polymers with high molecular weight and at rates comparable to the homogeneous plcoline-Cu complex. 2) Matrix isolation of catalytic centers affords 97

several advantages over homogeneous monomeric catalysts such as: prevention of agglomeration of metal, prevention of dimerization of highly active catalytic species, solution for inherent solubility limitations of a catalyst and retention of the coordinative unsaturated sites. Carbonylation of allyl chloride was catalyzed by [PdCNH^jj]* which was anchored to a cation 203 exchange resin. J At low catalyst concentrations, the homogeneous and polymer-supported catalysts had the same specific activity. As the concentration of catalyst was increased, the polymer-supported catalyst maintained constant specific activity, whereas the homogeneous catalyst became relatively less effective due to the aggregation of homogeneous catalyst to inactive complexes. It is well known that a basic requirement of homogeneous catalysis is the presence of coordinatively unsaturated sites. The rigidity of polymer prevents the recombination (bridging) of anchored catalysts and thus generates higher concentrations of unsaturated sites. A good example of this effect is the hydrogenation of 204 olefin in the presence of titanocene. Titanocene dichloride is reduced with butyl lithium to give a transient and highly reactive titanocene species which soon collapses to an inactive ditltanocene dihydridride. 98

This complication was overcome by anchoring titancene dichloride to a rigid styrene-DVB copolymer (2056 cross-linking) (119) followed by reduction with n-butyl lithium to a catalytic species (120) (equation 47) which is 16 times as active as the homogeneous analogue for the hydrogenation of cyclohexene.

L i + Cl Til ^C1

n-BuLi FT’Xl U? W) Ti t O ] 120

A series of studies on polymer-supported Wilkinson's catalysts show that the microreticular cross-linked styrene-DVB copolymer is still flexible enough to allow for chelation to occur to some extent. It required at least 2056 DVB cross-linking as the lower limit of macroreticular polystyrenes to assume that the metal centers do not interact due to the rigidity of the polymer network. Attachment of CpFe^O^H, which is very unstable and rapidly loses Hg to give the 205 dinuclear Cp2Fe2 (C0 )if, to polystyrene — 1 8% 99

DVB matrix generated a quite stable polymer - supported mononuclear species. The coordination number is lowered at the metal center (coordinatively unsaturated) by limiting the mobility of polymer-supported ligands. Soluble polymer-supported species behave like the monomeric complex. Phosphinated polystyrene cross-linked with 1? DVB was prepared at different P:Ir ratios and loadings employed for the hydrogenation of 207 1,5-cyclooctadiene in benzene. It was found that the polymer-supported catalysts were much more active at low P:Ir ratios (3 or 4) than those at higher P:Ir ratios, and when P:Ir ratio was 22 the rates of homogeneous and polymer-supported systems were about equal. The presence of 1% DVB was not sufficient to disrupt the mobility of the polymer support. 3) Multifunctional catalytic reactions. Multiple step transformations involving several metal catalysts have been developed and successfully applied in several cases. Polymer matrices allow the introduction of catalytic sites independent of each other. Simultaneous attachment of all necessary catalysts onto a single polymer support is an attractive approach to a sequential synthesis. All catalysts may be non-interacting due to the isolation effect of polymer matrices, but the substrate could migrate freely to each site. However, new questions arise as to the catalytic 100 systems. Will the catalyst react as it does individually? Will the presence of another catalyst interfere or change the reaction pathway promoted by the first catalyst? So far, there are at least four major sequential syntheses which have been put into practice. The sequential cyclooligomerization-hydroformylation of butadiene has been accomplished. The Ni and Rh catalysts can either be attached to the same polymer or to two polymers which are combined in the reactor. The reaction conditions are set initially for cyclooligomerization then adjusted to optimize on r hydroformylation (equation 48).

PPh2 )2Ni(C0)2 CHO PPh2 )3RhH(C0) +

The vinylcyclohexene generated by cyclooligomerization was selectively hydroformylated to give two isomeric aldehydes. The endocyclic double bonds remained intact. Sequential catalytic cyclooligomerization-hydrogenation 101

p n R reactions have also been performed. As shown in equation (49), butadiene was first cyclooligomerized by Ni sites followed by selective hydrogenation on Ru sites to monoenes.

pPh2 )2Ni(c°)2 0 C pPh2)2Ru(CO)2cl2

O-

(49)

The sequential dimerization-hydroformylation of butadiene with Ni and Rh catalysts represents another multistep process, as shown in equation^^ (50).

r py~pph2)?NiBr24 NaBH- r

f p V" pph2 ) 3RhH ( CO)

4 X 0 (50)

Finally, an interesting sequential hydroformylation-aldol condensation-hydrogenation was 102 performed by a single Rh catalyst in the presence of 20Q amino groups. 7 A styrene-DVB copolymer was modified with phosphine group, which later was complexed with Rh, then with amine groups(120).

^ N j ^ P P h 2-RhCl(CO) (PPh3)

n r 2 120

The propylene was hydroformylated with Rh-phosphine complex to butyraldehyde which underwent an aldol condensation catalyzed by the amine groups. The resulting unsaturated aldehydes were then hydrogenated by the rhodium complexes again to give primarily 2-ethylhexanal and 2-ethyl-4-methyl pentanal. In conclusion, the high cost of the homogeneoous catalysts makes recycling necessary in industrial production. The contamination of product by catalyst residues creates additional problems in purification process. Phase-transfer processes and ultrafiltration by menbrane-type reactors have been used to recycle the catalysts. However, it appears that the attachment of homogeneous catalysts to polymer support is the best way of recycling. Diffusion problems which appear to be the 103 main limitations, can be reduced by proper choice of polymer support and swelling solvent. The anchored catalyst often exhibits improved stability relative to its homogeneous analog. In fact, the formation of a new catalytic species may generate new catalytic pathways. The major problem associated with polymer-supported catalysts is metal-leaching. The activity of recycled supported catalyst is reduced due to the dissolution of metal into the reaction media to some extent. This is really a serious limitation to technical use. In some cases, this problem can be overcome by polydentate binding. This situation can be improved in going from liquid-phase catalysis to gas-phase catalysis. Finally, the durability of active species of polymer-supported catalysts photogenerated by irradiation may be further improved in photochemical processes. IV. Results and Discussion

4.1 Synthesis of Dipyridyl Derivatives for Immobilization on Polymers. Pyridines and dipyridyls were chosen as the candidates to replace Ph^P as ligands in palladium-catalyzed reactions, because: (a) stable organometallics bearing pyridines and, especially, with dipyridyls, as ligands are well known; Cb) closely related synthetic pathways are available or well-documented for the preparation of functionalized pyridyls and dipyridyls suitable for the formation of a linkage between polymer backbone and the ligand; Cc) the stability of pyridines and dipyridyls toward oxidation in the air; and (d) their basic character, which might cocatalyze Heck type reactions. Initially, extensive effort was devoted to the functionalization of dipyridyls. The first example is outlined in Scheme XIX.

104 105

SCHEME XIX 1)KMn04 ,£ MeLi/Ether I)HC1

121

© 4 X™ , 122 Oleum 20% ©J® HgSO^

Monomethylation of 2 ,2*-dipyridyl with MeLi in ether afforded liquid 6-methyl-2,21 -bipyridyl2^ (121). Attempted oxidation of 121 with KMnO^ failed to yield 6-(2,2'-dipyridyl)carboxylic acid (122). It was reported that when the same sequence was applied to 6,6,-dimethyl(2,2l-dipyridyl), 6,6,-(2,2,-dipyridyl)dicarboxylic acid2^ could be isolated. The failure to isolate 122 could be attributed to the high solubility of 122 in aqueous solution during the isolation process. Another approach, as shown in Scheme XIX, was the sulfonation of 2,2,-dipyridyl. Under extreme conditions (220°C, 24 hr), 2,2’-dipyridyl was allowed to react with 20$ HgSO^ and HgSO^ to afford 5-(2,2'-dipyridyl) sulfonic acid 123 in 83.4$ 106 yield. 212 However, 123 decomposed when dried in an oven and purification was accompanied by extensive decomposition. In Scheme XIX, 2-(2-pyridyl)cinchoninic acid (124) was selected because the functionalized dipyridyl derivative could be prepared in 94? yield in a single step, e.g., a cyclization reaction of isatin and 2-acetylpyridine in strongly alkaline ethanol solution (KOH/EtOH) ,2^ The second pyridine moiety was generated through a ring opening, cyclization, and condensation processes. A variety of approaches for the preparation of dipyridyls containing olefins and other functional groups via 2-(2-pyridyl)cinchoninic acid (124) and 2-(2-pyridyl)cinchoninoyl chloride (130) are outlined in Scheme XX and Scheme XXI, respectively. The acyl chloride derivative (130) was synthesized by treatment of 124 with excess SOClg. After removing the unreacted SOClg by vacuum, crude 130, a hard yellow solid, was obtained. It was found that the ethyl ester (125), which was the essential precursor to 127, can be synthesized either by acid-catalyzed esterification of 124 with EtOH/HgSO^ or by nucleophilic substitution of 130 with EtOH. The former is a clean and high-yield reaction, whereas, the latter is a sluggish reaction accompanied byy extensive by-product formation. When the ester 125 was subjected to nucleophilic attack by a SCHEME XX

The preparation of derivatives via 2-(2-pyridyl)cinchoni- nic acid (1££)

cnh{ch2 )3 © U $ 0 : zc-z ^7 i i I CH,Msl 4-

Polymer 125

CON(CH?),NH H2S0A . 1 H - 3 2 EtOH 109

Grignard reagent (CH^MgBr/ether), a mixture of the expected tertiary alcohol (126), ketone (131) and other unidentified impurities were obtained. Upon treatment of the mixture with p2®5 » evidence the presence of the isopropenyl derivative (127) was found by mass spectral analysis. In spite of several efforts to improve the purity and yield of 127, a satisfactory synthesis via this route was not achieved. The failure to prepare alcohol (126) from ketone (131) motivated an attempt to synthesize 131 directly. This ketone could be reduced to the corresponding secondary alcohol, which would yield a vinyl monomer upon dehydration. Carboxyylic acid can be reductively alkylated to methyl ketone through nucleophilic attack by MeLi and 21U subsequent hydrolysis. However the cinchoninic acid (124) failed to react with methyl lithium in this fashion. The feasibility of attaching dipyridyls to a preformed polymer backbone via a polyfunctional spacer was investigated. Treatment of the ethyl ester, 125, with refluxing 1,3-diaminopropane afforded an amino-amide (132) in moderate yield. The nucleophilic activity of the amino group of 132 was demonstrated by generating a sulfonamide (133) from a coupling reaction with p-toluenesulfonyl chloride. Thus 108

SCHEME XXI Preparation of derivatives via 2~{2-pyridyl)cinchoninoyl chloride(150)

cg(CH2)55l- H ;o2Et A O D

EtOH

nch2-ch=ch2

QUQQ g u g g

0CH2CH2 0H PhCH20H NaN,/H20

0 ™ 2 ™ 2

* 135 139 NaOH/H2o[ mh2

o -OQ

Polymer I 1r-\ | - O * s o 2ci

137 110

SO^Cl-containing polymer is a promising precursor for the preparation of a dipyridyl-containing polymer. An attempt to prepare amine 136 via a Curtius 215 rearrangement was made. An acyl azide, prepared by the reaction of acyl chloride 130 with sodium azide, was converted into an isocyanate (135) upon warming the crude azide in toluene. Isocyanate (135), identified by M.S. and IR, was a stable brown solid. Subjecting 135 to basic hydrolysis conditions (Na0H/H20), only a minor amount of amine 136 could be detected. Neither 135 nor 136 could be isolated in pure form. Benzyl ester (129) was chosen as model compound for a polymeric dipyridyl derivative based upon a polystyrene backbone. The ester could be prepared from either 124 or 130. Since the acyl chloride did not appear to exhibit normal reactivity, the carboxylic acid (124) was esterified directly. Treatment of the sodium carboxylate salt 128 in HMPA with benzyl chloride effected benzylation of 128 to afford 129 in moderate yield. Under similar conditions, vinyl benzyl chloride reacted to yield a styrene derivative 134. This synthesis proved to be the most direct route to a vinyl monomer with a bipyridyl substituent. Three additional dipyridyl-containing olefin moieties, e.g. 138, 139 and 140 were successfully 111

synthesized. However, the successful synthesis of the vinyl benzyl ester (13*0 precluded further investigation of their polymerizability. 4.2 Synthesis of Pyridine Derivatives for Immobilization on a Polymer Support, The successful synthesis of a monomer containing dipyridyl moiety motivated an extension to a series of similar pyridine derivatives with an aim to study electronic effect of pyridine substituents on the catalytic activity of palladium compounds. 4-Pyridinecarboxylic acid (141), was prepared from the oxidation reaction of 4-picoline with KMnO^ in 48.1% yield.216 (Scheme XXII)

SCHEME XXII

KMnO

H Cl

To synthesize a model compound benzyl 4-pyridinecarboxylate (143), nucleophilic substitution of benzyl chloride by sodium 4-pyridinecarboxylate 112

(142) was the preferred approach. The tendency of high-boiling benzyl alcohol to undergo acid-catalyzed condensations at high temperature precluded the ordinary acid-catalyzed esterification of the carboxylic acid (141). R.C. Larock presented a feasible room-temperature esterification of benzyl chloride with sodium acetate in HMPA to afford 93? yield of 217 ester. ' Similar reaction condition was followed to produce the esters shown in Scheme XXIII.

SCHEME XXIII

:0oH HMPA.RT 2 days 142

Due to the competetive formation of water-soluble quaternary ammonium chloride salts 144 and 146, the desired esters were isolated in only 60? yield. Monomer (145), is only moderately stable In air, however the monomer could be stored under nitrogen. Scheme XXIV shows different approaches to the preparation of a pyridine derivative with a strong electron-donating substituent in the 4-position,

4-benzyloxypyridine (147). 113

SCHEME XXIV

PhCH2Cl CH2ph

0-+Na ICH-Ph PCI 154 Na o 0 147 148

PhCH-Cl HaNO, RT

Ha

HMPA

155 151 Following an analogous procedure for the preparation of 143, the alkoxide (152) was prepared from 151 and treated with benzyl chloride in HMPA to afford 147 only in 20% yield, due to the strong electron-donating character of alkoxide (152), (resonance form of (153)) which produces a high electron density on nitrogen. The predominant site for nucleophilic attack is nitrogen and a water soluble quaternary ammonium salt (154) is formed. A higher yield (36?) of o-alkylated product was obtained when 3-hydroxypyridine was benzylated under a similar condition. Two other approaches, e.g., 148 to 147 and 150 to 147, where nitrogens were protected as nitrogen oxides, were examined. However, the benzylation of 150 failed to give 149; the predominant benzylation appears to occur on the oxygen of nitrogen oxide. However, 149 was successsfully prepared in good yield from 4-nitropyridine 1-oxide upon treatment with excess sodium P18 benzyloxide. When 149 was subjected to N — > 0 deoxygenation by PCl^, ^ 7 was isolated in fairly low yield. Thus, attempts to improve the synthesis by protecting the nitrogen were not fruitful. Monomer, 155, was prepared by the most convenient one-step synthesis, i.e., 151 to 147. A corresponding attempt to prepare 4-benzylamino pyridine as shown in Scheme XXV 115 was not successful.

SCHEME XXV NH, NHCH2Ph NaOK or ^ N a + 0 ( ) o N NaNH? £

However, it was reported 2-benzylaminopyridine was prepared from the reaction of 2-aminopyridine, KOH and 21 9 benzyl alcohol in good yield. 4.3 Homogeneous Palladium-Catalyzed Vinylic Substitution. 220 In a typical Heck reaction (equation 51), Ph^P is used as the ligand to activate and stabilize the palladium catalyst. Triethylamine is required to neutralize the acid generated and to regenerate the precatalyst Pd(o). When X = I, Ph^P is not a necessity. DMF, a weakly-donating ligand, activates the Pd catalyst and coupling is effective; however, the formation of Pd black Is always the inevitable result if Et^N is employed as the base. 116

EKF,Ne2PdClA (1mol. #)

* O ^ Q 6 - PPh,, Et N, 90*c 3 3 * EtjNHX (51)

In order to improve the catalytic system and preserve the Pd catalyst as homogeneous catalyst as long as possible, different combinations of ligand, base and solvent were evaluated. If an optimum ligand, catalyst and solvent combination could be found, the stable homogeneous catalyst would facilitate kinetic studies on the catalytic system. Thus, further insight on the mechanisms of Heck reaction would be possible. When the triphenylphosphine ligand is replaced by 4-picollne in equation 51, where X = I, Pd black precipitates In a short time. If triethylamine is used as the base, it is obvious that U-picoline is coordinated to Pd(O) complex more weaklyy than Ph^P since no Pd black precipitation was observed in the presence of the phosphine ligand. We found that the nature of the base, under these circumstances, appeared to be a key factor in determining the fate of the Pd(0) complex. A series of tests using a wide variety of bases revealed that proton sponge and morphollne are the two most effective bases when 4-picoline is the ligand. 117

Morpholine was selected as the base, because the high cost of proton sponge would make it impractical on a large scale catalytic reaction, and also the precipitation of the proton sponge hydrochloride salt during the reactions makes the stirring extraordinarily difficult. When bromobenzene (equation 51, X = Br) was used as the substrate under the similar conditions to those used for iodobenzene, the catalytic reaction failed. Table X summarizes a wide variety of conditions where catalytic vinylation reactions were attempted with bromobenzene and styrene. When tertiary or secondary bases were used in the system, Pd black precipitated in almost every case (entries 1 to 12) except in entries 1, 2 and 6 where primary amines were used as ligands. A very stable complex formed between primary amines and Na^PdCl^, which produced an Inactive Pd species. In entry 9, reaction occurred to a small extent but Pd black was precipitated. Triphenylphosphine was a good ligand with which reaction proceeded well (entry 12). Sulfur compounds were generally very poor ligands. In every case, no reaction was observed at all with the precipitation of Pd black (entries 13 to 16) except that diethyldisulfide can stabilize the Pd complex from decomposition (entry 16). When 4-picoline was used as 118 ligand (entries 18 to 24), none of the bases employed were able to effect the catalytic reaction. In entries 25-36, 2 ,6-diaminopyridine, a strong coordination ligand owing to its chelating properties, did not improve the catalytic reaction. The sole exception, entry 35, where a co-base system NaHCO^/morpholine effected the reactions in good yield, revealed the secret to success. The addition of catalytic amount of morpholine to act as a messenger between solid NaHCO^ and hydridopalladium bromide (157) was required (Scheme XXVI). The same activity was also observed when 4-picoline was the ligand (entry 37).

TABLE X Associated effect of ligand, base and solvent on the reactivity and stability of Pd complex for reaction of Heck type recation ( Bromobenzene is used ).

Entry Ligand Base Solvent Observation

1 h2 n(ch2 )3nh2 Et N IMF vhitel N.R. 3

t| M 2 ^n(ch2 )2 nh2 vhitel N.R.

3 n n Pd| N.R.

u 4 @-0 ti PdJ N.R. / N ( C H j )2 j^n(ch3 )2 5 ti N.R. ^ n(ch 3 )2 ^ N ( C H 3 )2 6 hn(ck2 ck2 kk-)2 0 « vhitel P< 4 ^ n ( c h 3 )2 H 7 »r N Pd i N.R. ^■m(ch3 )2 cont1d 119

6 Pd 1 N.R.

9 Pd 1 H,r» NH,

10 thick| N.R. ho2cJQW salt

11 Pd j, N.R.

12 PPh3 It reacts well< Small Pd4

13 Ph-S-Ph DMSO Pd i

14 ( C ^ S Pd i

15 EMSO O VNo Pdj soon 16 EtSSEt 6 DMSO stable N.R. 17 PhAGe EMSO p4 soon 0 H 18 6 no solv. 19 0 DMF 20 proton sponge

21 Et3N soon

22 O stable N.R. Y

23 stable 9 white J, 24 Pd J, N.R. '-C 25 stable N.R. NH-

26 h2n(ch2 )6 nh2 stable small white^ N.R. Table X cont’d

2 2 3 2 27 h n(ch ) nh " stable N.H.

26 'v'n(ch2ch?)nh2 " stable N.R.

PdJ 29 ~ ^ 0H

30 r~< DMF Pd J,

31 NH; o

32 proton ir sponge

33 f 1 /NaHCO, HMPA N . 3 H 34 polymer base *■ ", It (TBA) reacts -0>, 35 /NaHCO, o CO tit H * 'reacts 36 NaHCO stable H 3 slow reaction N- 37 /NaHCO, co2tIt ‘reacts well

PdJ: Pd black precipitates off

N.R.: NO significant reaction was observed in 12 hrs at 90"c.

J, : Precipitate

Stable: No Pd|is observed. Reaction condition: StyreneflO mmol.), bromobenzene(10 mmol.), ligand( 3 mmol.), Na2Pd£U( 0.1 mmol.), base(11 mmol.),solvent x ml. Total volume *8.54 ml at 90*c. 121

SCHEME XXVI

L H-Pd-Br ♦ L

157

NaHCO. 3 M I

( regenerated) +

Although the mechanism of decomposition of Pd complexes is not clear, one contributing factor may be the accumulation"of morpholine hydrobromide salt (159) in solution. Thus, concentration of intermediate (157) could be increasing because of an equilibrium due to the basicity of the Pd(0) species (158). Employment of

NaHCOg as a co-base will consume 159 and regenerate morpholine along with an evolution of C02 gas. This process prevents the accumulation of 157 and 159. In order to enhance the solubility of NaHCOg, HMPA was used as the solvent in this specific situation. 122

4.4 Substituent and Concentration Effect of Pyridine and Dipyridyl. Regardless of whether PdClI) or PdCO) complex was added as "precatalyst", the true catalytic species is believed to be the coordinatively unsaturated species of L 2 Pd(0) or LPd(O) (161) in Scheme XV. Scheme XV

Substrate 2L L2Pd(II) — LjjPd(O) — L^PdCO) — L2Pd(0) — LgPdCO) Substrate

160 161

L = Pyridine or Dipyridyl

At least, this is true when L = Ph^P,22^ and it appeared true for L = pyridine. Based on that assumption, any factor that favors the formation of 161 will also activate the catalyst; conversely, strong ligands may decrease the activity of the catalyst by enhancing the formation of 160. Molar ratios of 4-picoline to Pd (NagPdCl^) were varied in the range of 10 to 140 (Figure IV). Both Phi and PhBr were studied. For the purpose of comparison of the catalytic reactivity, the initial rate of a kinetic curve is utilized as an index of catalytic 123 FIGURE III The dependence of consumption rate of styrene vs. time on the ratio of 4-picoline to Na2PdCl..

80

70 / Pd

60 10:1

60:1

120:1

o 40

o 30

20

10

1 2 3 45 6 7 8 (hr.) Initial rate( 10“-* M*sec 20 18 10 2 - 10 ioiev. Pd vs. picoline Dependence of catalytic reactivity on the ratios of 4- of ratios the on reactivity catalytic of Dependence

20 0 0 0 0 0 0 0 1 10 3 4 10 1^0 150 140 130 120 110 100 90 80 70 60 50 40 FIGURE V I / NagPdCl^tMolar ratio; /NagPdCl^tMolar A : Phi A

124 125

reactivity. The kinetic curve is generated by plotting aryl halide or styrene consumption vs. reaction time. (Figure 3.) It was found that catalytic reactivity was inversely proportional to the molar ratio of 4-picoline to Pd in both cases (Phi and PhBr). Also the degree of dependence of catalytic reactivity on the variation of molar ratios (4-picoline to Pd) were roughly similar; nonetheless, Phi was more active than PhBr. The negative slope of Figure IV, essentially, supports the idea that the coordinatively unsaturated species (161) was required to effect a catalytic reaction. Generally, the catalytic reactivity was not extremely sensitive to the variation of molar ratios of 4-Picoline to Pd. The catalytic activity in the presence of 2,2l-dipyridyl, a strong chelating ligand, which is known to form a variety of Pd(II) complexes, was very sensitive to the variation of molar ratios of 2,2'-dipyridyl to Pd, even in the range of 1:1 to 3:1 shown in Figure V. This observation supports and implies the concept that coordinatively unsaturated species (161) are necessary for activity; thus, dissociation of one pyridine unit 162 — > 163 (Scheme XXVII a) is contributing substantially to the rate-determining step. Without the constraint of the bidentate chelating effect, the conversion of 164 to Initial rate ( 10 M-sec‘ 10 11 2 1 3 4 6 5 7 8 9 Dependence of catalytic reactivity on the ratio of 2,2 - 2,2 Pd of vs. ratio the on dipyridyl reactivity catalytic of Dependence 1 IUE V FIGURE 2 / N^PdCl^ ratio) (molar /N^PdCl^ 3 : Phi PhBr

4 126 Pd— ^ 7~>Ph 128

165 (Scheme XXVII b) proceeds much faster when the monodentate pyridine is the ligand. Based on the above observation and argument, any change that facilitates the dissociation processed (i.e. 162 to 163, 164 to 165) will enhance the catalytic activity, and vice versa. A further investigation on a variety of pyridine derivatives was performed as shown in Table XI. The relative catalytic reactivity is in decreasing order = 2-chloro > 4-cyano > M-COgCHgPh > 2-methyl > 3-hydroxyl > 2-methoxy > 4-hydroxyl > 4-methyl > 3-OCHgPh > 2,6-diamlno >> 4-amino. In general, electron withdrawing groups such as -Cl, -CN and -COgCHgPh enhance the catalytic activity as compared to 4-methyl. The bonding between nitrogen and Pd (164) is weakened due to the withdrawal of unshared electron density on the nitrogen atom through the inductive effect. On the other hand, electron-donating groups such as -OCHgPh and -NHg show a decrease in catalytic activity. In addition to the electron-donating character of the hydroxyl group, strong hydrogen bonding is formed intramolecularly between nitrogen and -OH, which also makes nitrogen less available for complexation (3-hydroxyl and 4-hydroxyl). It's noteworthy that there is a significant drop in catalytic activity in the 4-amino case. A very tight 129

TABLE XI Effect of Pyridines on the reactivity of vinylic substitution

Pyridines Initial rate(10“i*M'aec“'')

2.31 aci

2.28

02,CH2Ph

2.27 a 2.00 OH 1.98

1.88 O l OCH, iH 3

1.74

1.73

OCH„Ph a 1.51 jQ l 0.91

0.20 130

bond forms between primary amino groups and Pd, resulting in the formation of white precipitate. In addition to the electronic effects mentioned above, it is discerned that steric hindrance also plays an important role in the activation of the catalyst, e.g., 2-methyl > 4-methyl, 2-methoxy > 3-0CH,,Ph, by enhancing the dissociation of 164 to 165. It is appropriate to cite a specific study using benzyl 2-(2-pyridyl)cinchoninate (129), an analogue, of 2,2'-dipyridyl. In Figure VI, system (A) which includes 129 is more reactive than a system (B) based upon 2 ,2'-dipyridyl by 1.53 fold. Owing to both the electronic and steric factors, the fused pyridine was, presumably, able to undergo a more facile dissociation process compared to that of 162 to 163. The'slopes of the curves in Figures IV, V and VI are summarized in Table XIa. The values are approximately proportional to the bonding strength of nitrogen-Pd bond. The nitrogen-Pd bond strength have been correlated with Pd-Cl and Pd-N stretching frequency in a far infrared spectral study on a variety of 222 palladium compounds of nitrogen ligands. 4.5 Substituent Effect of Aryl Halides. In order to expand the scope and to test the functional group tolerance of palladium-catalyzed 131

TABLE XIa Change of slope of initial rate vs. ligand concentration on the variation of nitrogen ligands

Catalytic system PhX |Slope|

0.067 JL Phi 0 /NagPdCl^ PhBr 0.052

Phi 0.22 /Na2PdClA 0-0 PhBr 0.17 COoCHpPh ^ JL2 GCCO/Na2PdCl4 Phi 0.22

TABLE XII Substituent effect of Aryl iodides

Aryl iodides Initial rate(10“Sl-sec-1)

0CH3 2.13 - o -

1.60

1.37 - o ,ci

I 0.7* Initial rate (10--JH-Bec 20 19 17 18 15 16 13 12 1*» 10 11 6 7 6 9 2 h - 5 3 ‘

The catalytictstudy of using benzyl 2-(2-pyridyl)cinchoninate benzyl of using catalytictstudy The (129) and (129) and 2,2 -iyiy. a^CHjjPh -dipyridyl. FIGURE VI Dipyridyl / NagPdCl^ (molar ratio) _L 2 3 _L 132

133

vinylic substitution using 4-picoline as ligand, the reactivity of various substituted aryl iodides and aryl bromides was investigated. The relative initial rates observed are shown in Table XII and XIII respectively. For aryl iodides, the relative reactivity is in the following decreasing order: 4-methoxy > 3-methyl > -H > 3-chloro. . For aryl bromides, is 4-methoxy > 3-dimethylamino > 4-methyl > 3-methoxy > -H > 2-methyl > 4-amino > 4-formyl > 4-trifluoromethyl > 4-cyano > 4-nitro. The reactivity of iodobenzene is parallel to bromobenzene in terms of substituent effect. The reaction clearly was accelerated by electron-donating substituents on the arylhalides. This observation is opposite to that reported for the palladium-catalyzed 22 223 arylation of ethyl acrylate with aryl halides. The relative rates obtained for various para substituents have been plotted against the Hammett P values (Figure VII). Two linear correlations can be drawn, one for electron-withdrawing and the other for electron-donating substituents. The P value for the former (P - -1.20) is more negative than that for the latter (P - -0.76), indicating higher sensitivity to substituent effects for electron-withdrawing groups. 134

TABLE XIII Substituent effect of Aryl bromides

Aryl bromides Initial rate {10"4H .sec-ij

CH3 ° - © ' Br 1.95

2N- < Q > - Br 1.63

1.43

CHjO

1.43

1.35 0 Br

1.22 ^ HBr

1.14 H2N " ^ ^ " Br

Q H C - ^ ^ ^ r 0.41

0.22 F3 ° ^ © ^ r

0.17

too slow °2N “ ^ ^ ' Br log F(x)/F(H) 0.9 -0 0.00 -0.3 -0.5 - -0.7 - - - -0.4 0.3 0.2 0.1 0.1 0.2 0.6 0.8 Plot of Hammet equation for substituent bromidesAryl effectof equation for of Hammet Plot p-OCH FIGURE Vn P-CN

'-CF- 135 136

Recall that the mechanism proposed by Heck included oxidative addition and phenyl migration (Scheme XXIX).

SCHEME XXIX , PdX

Oxidative Complexation + Pdt0j addition* ' * ■i*1

Phenyl

migration

According to a study by Fitton for the oxidative addition of substituted aryl halides to [(Ph^Pl^Pd, electron-withdrawing group increased the rate. For Heck-type reactions, when ethyl acrylate was used, the substituent effect was analogous to that of Fitton*s op study, indicating oxidative addition Is the rate-determining step with a facile phenyl insertion to methyl acrylate, where > k^. However, in our study, when styrene (R* a Ph) was employed a reverse substituent effect was observed, implying (k^ > kg) phenyl migration rather than oxidative addition is the rate-determining step. The order of alkene activity has been established in the reaction of phenylmercury (II) chloride catalyzed CO by LigPdCljj with olefin, p as below. 137

CH2=CH2 > Me02CCH=CH2 > MeCH=CH2 > PhCH=CH2 > PhMeCh=CH2

14,000 970 220 42 1

The steric hindrance dominates the reaction rate When a change was made from methyl acrylate to styrene, a rate drop by ~ 25 fold was noticed which accounts for a swap of rate-determining steps. The main factor that controls the reactivity of phenyl insertion was,

presumably, the bond strength of Pd-Ph. Basically, the bonding between Pd and phenyl consists of one

d *R

<7-bonding is not very sensitive to the substituent effect, whileas, the rt-bonding is the resultant of back donation of palladium d electrons (HOMO) to a 234 vacant n* orbital of the aromatic ring. Electron- withdrawing group CR group) increases the TL-acidity of 7l * orbital and results in strengthening the Pd-Ph bond owing to a strong back donation of d electrons of the Pd center. The stronger the Pd-Ph bond is, the slower the 138 phenyl migration (insertion) will occur. Several isolated stilbenes prepared from palladium-catalyzed arylation of styrene with substituted aryl bromides are listed in Table XIV. All substituted stilbenes are in E-form Identified by the comparison with literature melting points. One case was studied extensively In which the reaction mixture of bromobenzene and styrene was analyzed by HPLC. No cis-stilbene was detectable, only trans-stilbene was formed by comparison with authentic cis- and trans-stilbenes. As shown in Table XIV, trans-substituted stilbenes have markedly different melting points from their corresponding cis-isomers which are usually liquid at R.T. 139

Table XIV Isolated styrenes from palladium-catalyzed arylation of styrene with substituted aryl bromides.

Entry Compounds Reaction Conversion0 m.p. C B.P. °C Ref time (hr) (%) (trans) (nmHg) (cis)

1 4-Methoxystilbene 2.0 76.5 135-136 98-100(2) 227 (136) 226, 228

2 N,N-Dimethyl-4- - 1.5 66.0 149-150 28-9f 229 stllbene (151-152) 228

3 4-Methylstilbene 1.5 63.0 119-120 105-110(2) 227 (119) 226 4 Stilbene 3.5 68.5 122-124 82-84(0.4) 227 (124) 226, 228 5 4-Stilbeneamine 3.0 57.0 140-143 (144) 228

6 4-Trifluoro- 20 69.5 133-134 stilbene (132.6-133.6) 226

7 4-Chlorostilbenee 3 1/6 86.0 127-128 108(1) 225 (128) 228

a. All reactions were run in HMPA, NaHCO,/ morpholine used as base at 90 C. b. All compounds isolated are in E forms. c. Conversion percentages were obtained by G.C. analysis. The disappearance rate of styrene vs. time was monitored. p-Xylene was used as the internal standard. d. Boiling point (mmHg) for cis isomer reported In the literature are listed. e. Prepared from p-chlorostyrene and iodobenzene. f. Melting point. 140

4.6 Polymer-supported catalyst. Three monomers 134, 145 and 155, synthesized according to Scheme XX, XXIII and XXIV respectively, are listed in Table XV.

Generally, the yield of the desired isomer was determined by the nucleophilicity of nitrogen, which is controlled by the electronic effect of the substituent on pyridine unit. Electron-withdrawing groups decrease the electron density on nitrogen and thus increases the yield of the oxygen adduct (e.g. 134, 145). Electron-donating groups such as alkoxide (155) enhance significantly the nucleophilicity of nitrogen resulting in only 18% of isolated yield of 155. The side product was, in every case, presumed to be a water-soluble quaternary ammonium salt. Since commercially available vinylbenzyl chloride is in fact a mixture of almost equal amounts of p- and m- isomers, 141

TABLE XV Monomers synthesized for polymer-supported catalysts

physical Monomers yield data Elemental analysis

c HN p-isomer 38# 78.67 7.■ 65 QLOQ 4.-95 78.45 7..57 4..93 134

semisolidt C HN 57# a mixture of p- and m- 75.30 5.,85 6 isomers 5.,48 75.23 5..74 145 5-,47

OCH. viscous liquid C HN 18# a mixture of p- and rn- 79.59 6 ..63 6' isomers 6 .,20 79.37 6 .,54 155 6 .16

For IR, NMR, and MS, they were listed in Experimental part. 142

isolation of pure monomers was difficult. The crystalline p-isomer of 134 was purified, but separation of p- and m-isomers for 145 and 155 was unsuccessful. Column-chromatographed mixtures were used for subsequent polymerization process. The monomers are not very stable in the air, but could be stored under nitrogen or used right away. Several divinylbenzene (DVB) cross-linked polymer beads were prepared (Table XVI) using suspension polymerization techniques.Poly(vinylalcohol) was used as a stabilizer for the formation of beads, toluene and H20 (1:10) together give a suitable solvent system, and AIBN was used as an Initiator. The polymerization was complete, i.e., > 95%, conversion within 24 hrs at 70°C (equation 52).

Poly(vinylalcohol), HgO

>

Monomer + AIBN, Toluene, 70°C Ng flushed, 24 hr.

(52) TABLE XVI Polymers prepared for Polymer-supported Catalysts

Entry polymer Monomer mequiv. of N/g of ______polymer a © - O "N CD 90# 10# 8.27 85# 15# 8.01

Q, 1 » O

(II) 90# 10# 3-60

& d 3.80

(III) 90# 10#

4

(IV) 10# 10# 80# 4.40

5 Q g

10# 10# 80# 1.57 (V)

All the polymers above were prepared according to equation 10. a.Mequiv. of nitrogen per gram of polymer was calculated from Nitrogen contents obtained from elemental analysis. b.Entry 1,2 and 3 were in monomeric ratio by weight, entry 4 & 5 were in monomeric ratio by mole. 144

The corresponding palladium dichloride complexes of polymer containing pendant nitrogen ligands were prepared according to equation 53.

NagPdCl^/EtOH

d c m ------> (p>0°2Pecis / R.T. Stir for 1 day v--/ 2 Swollen Polymer

The polymer support was swollen in DCM for 1 hr, to which a homogeneous ethanol solution of NagPdCl^ was added. After stirring for 24 hr at R.T., yellow or pale yellow polymer supported catalyst was isolated and washed with DCM and ethanol alternatively. The filtrate was analyzed by a colorimetric method where the absorption of a palladium p-nitrosodimethylanillne complex was measured at 525 mu. with a UV-vis. spectrophotometer. J A negligible amount of

NagPdCl^ (< 10“6g) remained in the filtrate, when the ratio of N:Pd is larger than 7:1 for polymer I, II and III. A typical Heck type reaction, vlnylation of aryl iodide with styrene, was run using polymer-supported catalyst. The results were summarized in Table XVII. 145

TABLE XVII Catalytic reaction of Heck type reaction by Polymer-supported catalysts Initial Reactivity Polymer catalyst PhX Cycle rate decayed

m__i y (10"Sl sec-1)

1096 Phi 1 1.63 (VI) 2 1.44 68.3# 15# 1 1.81

2 1-35 74*5#

3 0.79 43-6#

c h 2o )2PdCl2

(VII) 3-96 Pd|

0CH2«JN)2pdcl2 „ 1 1.12

(VIII) 2 0.95 84.8# 3 0.66 58.9#

2.76 PdCl (XI) Too slow

The ratio of N/Pd for Polymer catalysts (VJ-(VIH is 30/1, for ( X I ) is 10/1 146

In the first cycle, polymer-supported (VI) and homogeneous systems exhibit similar activity (for -4 homogeneous reactions, the initial rate is 1.73 x 10

M. sec“^ when Phi was used). No significant difference between 10$ and 15$ divinylbenzene cross-linked polymer catalyst was observed. 10$ and 15$ of polymer-PdClg complex (Polymer VI) have comparable swellability of 20$ in HMPA at R.T. The PdClg was envisaged to be complexed mainly on the surface of polymer supported and thus the catalytic reaction was not related to the degree of crosslinking of the polystyrene in the range of 10$ to 15$ crosslinking. Only 74.5$ and 43.6$ of the initial polymer support catalytic reactivity were retained for 2nd and 3rd recycles respectively. After each cycle, polymer-supported catalyst was isolated and washed with small quantity of DMF, and then fresh reagents were added to run the subsequent cycle. Colorimetric analysis for Pd in the wash solvent revealed that there was an obvious leaching of Pd, which accounts partly for the decayed reactivity of reused polymer-supported catalyst. Both the DMF solvent and precipitated stilbene product were found to be contaminated with palladium. When PhBr was used, a NaHCO^ slurry was the required base. Even in the first cycle, a 147 relatively slower initial rate results from the incomplete mixing of a slurry of both polymer supported catalyst and base. For polymer supported catalyst VII, an extraordinarily fast initial rate (3.96 units) in the first cycle accompanied by the Pd black precipitation was amazing. A similar rate enhancement at the 1st cycle with the formation of Pd black was also noticed when polymer supported dipyridyl-PdClg catalyst XI was employed. However the 2nd cycle of reused catalyst' XI showed a marked decrease in activity, which indicates another catalytic deactivation process, e.g. the initially formed highly active Pd metal atom collapses into a much less active Pd agglomerate. On account of the strong electron-donating character of the 4-oxopyridine moiety, polymer catalyst VIII showed a slower inittial rate (1.12) than that obtained with systems derived from electron poor pyridine systems. The initial rates dropped to 84.8% and 58.9% of that of the 1st cycle for 2nd and 3rd cycle respectively. The palladium leaching problem was less serious for VIII than for VI, probably because the stronger nucleophilicity of pyridine tends to bind the PdClg more effectively. The palladium leaching problem remains, nevertheless, quite serious. Based on the above study, a possible improvement 148 was made to modify the existing polymer catalytic system. Xylene, a nonpolar and very poor coordinating ligand for Pd, was chosen to replace DMF. A low nitrogen-containing PVP copolymer (V) was prepared to improve the swellability of polymer backbone. Considering the poor solubility of pyridine-PdClg complex in xylene, namely, less dissociation of pyridine off pyrldine-PdClg complex, a higher ratio of N:Pd (8:1) was employed in polymer catalyst. Heck type reaction was then investigated and the results were described in Table XVIII. The initial rate, based on the same N:Pd ratio (8:1) was significantly slower (initial rate = 1.09) for polymer catalyst X in xylene than the corresponding polymer catalyst in DMF (initial rate = 1.63* for polymer VI at N/Pd = 30). The inherently low solubility of pyridine-PdCl2 complex in xylene probably accounts for this low reactivity. A more serious problem is presented by the precipitated morpholium hydrobromide which is insoluble in xylene, and tends to coat and block the catalytically active center. When the co-base system, NaHCO^/morpholine, was used, a resulting much slower rate indicated that the morpholine could not function as a messenger due to the insolubility of the intermediate hydrochloride. Water washings from the reaction mixture containing polymer catalyst (X) TABLE XVIII Solvent effect of xylene for Heck type reaction

Initial (10-4 Polymer catalysts Base N/Pd rate M sec"']

©-Oh)2PdC12O 8:1 1.09 (X) H

I"0} /NaHCOj 8:1 0.11 @ - @ i 0 2PdCl2 H

( E > @ ,>2pdcl2 O 30:1 0.22 tj (VI)

0 / Na2PdCl4 30:1 0.89 0N H In every case, Phi was used. Other conditions were similar to Heck type reactions (equation 1). 150

contained palladium as indicated by a palladium black precipitate when the solutions were treated with KaBH^ in ethanol. The Pd leaching remains a problem. When higher ratio of N:Pd (30:1) was tested (polymer catalyst VI), a very slow initial rate suggests that generation of the coordinatively unsaturated catalytic species is inhibited by the non-polar environment at higher ratios of nitrogen to palladium. Based on the above study, DMF is a better solvent than xylene because morpholium hydroiodide is soluble in DMF and furthermore it facilitates the dissociation of ligand to afford the coordinatively unsaturated catalytic species. In order to understand the palladium leaching process when xylene was used as solvent, a series of stability tests were conducted. The results are summarized in Table XIX. Polymer supported catalyst (VI) was stable in the presence of H^O, xylene or

HgO/HaHCO^, but unstable when either styrene or morpholine were present. Both styrene and morpholine will coordinate weakly with PdCl2 and through this coordination palladium leaches into solution. 151

TABLE XIX Metal leaching of polymer catalyst(VI) in the presence of substrates (Tests of filtrate) Run Substrates Observation(NaBH^/EtOH)

N.R. No PdJ, H 2°

Styrene Pdl , polymer turns grey

morpholine Pd|

p-xylene No PdJ,

bromobenzene, No Pd| p-xylene

H20,NaHC03 No PdJ

styrene, ELO PdJ, bromobenzehe, NaHCO^

0.1 g of polymer catalyst(VI) and adequate amount of each tested substrate were added and heated up at 90*c for 12 hrs. Polymer was isolated, the filtrate was treated with NaBHA/EtOH. 152

4.7 Conclusions In conclusion, for Heck type reactions, 4-picoline may be utilized in place of triphenylphosphine. The former proves to be a better ligand than the latter for the process can be run in an open vessel rather than in a sealed system under Ar. Through a series of studies on ligand concentration effects, ligand substituent effects, and substituent effects on aryl halides, more progress was made on the elucidation of the mechanistic aspects of catalytic cycle. It was found that selection of the proton acceptor used to neutralize hydrogen halide generated from coupling reaction was the key step in effecting the reaction with aryl bromides. Several nitrogen (pyridine or dipyridyl)-containing polymer supported palladium dichloride catalysts were prepared and subjected to Heck type reactions. In every case, comparable reaction rates to homogeneous systems were observed for the first cycle, but a significant decay of catalytic reactivity was found at 2nd and 3rd recycle of polymer supported catalyst. Possible reasons for the reactivity decay are: the palladium leaching into bulk solution through a weak association (or coordination) with styrene, morpholine and DMF or HMPA. Further, palladium catalyst will decompose and form inactive Pd agglomerate without coordinating to a stronger ligand. 153

The use of non-polar solvents is feasible if a hydrophobic polymer matrix is employed but the Pd leaching problem remains unsolved. It is reasonable to assume that pyridine or 2,2'-dipyridyl may replace triphenylphosphine and serve as ligand for palladium or other metal-mediated (catalytic) reactions, if optimum conditions are found. Through the comparison of results from ligand systems (N and P), a better understanding on the mechanistic aspects of catalysis may be achieved. For polymer-supported catalysis, in order to improve the catalytic recycle efficiency, a deliberate selection of adequate catalytic reaction system is necessary. A system without involving the dissociation of binding ligand will be the most promising. V. Experimental Section

General Information Nuclear magnetic resonance spectra (NMR) were obtained using a Varian Associates HA-60 Spectrometer. 1H NMR chemical shifts are reported in parts per million (ppm,£) downfield from tetramethylsilane (TMS), and the usual notations were used to describe the spectra; s = singlet; d = doublet; dd = pair of doublets; m = multiple; b = broad and J = coupling constant. Infrared spectra (IR) were recorded on a Perkin-Elmer Model 621 Spectrophotometer. Mass spectrograms were obtained on a Hewlett-Packard 5895-GCMS at 70 eV. Gas chromatographic analyses were carried out on a Hewlett-Packard model 5700A using a carbowax 20M, 8' x 1/10" A1 column and helium as a carrier gas. p-Xylene was used as internal standard. Column chromatography was performed on activity-1 alumina or silica gel. Thin layer chromatography was performed on commercially available TLC plates. The elemental analyses were performed by Mic-Anal

(Tucson).

154 155

Melting points were determined on a Thomas-Hoover capillary melting point apparatus and were uncorrected. Commercial reagent grade solvents were utilized without further purification. Dimethylforamide (DMF) and hexamethylphosphoroustriamide (HMPA) were dried over activated molecular sieves (Linde Type, 4A). Reagents used were either purchased or synthesized as described. Colorimetric determination of palladium follows the analytical methods for noble metals by R. Belcher and L. Gordon. Preparation of 2-_(2-Pvridvl)cinchoninic acid (1g*P..213 In a 250 ml round-buttomed flask 7.3 g (49.6 mmol) of isatin was dissolved in an alkaline solution containing 152 ml of 50% aq. ethanol and 49.6 g KOH. To the resulting ice-cooled deep purple solution of 2-acetylpyridine (6.0 g, 49*6 mmool) was added slowly with good stirring, and the solution was refluxed gently for 4 hrs. The ethanol was removed by simple distillation. The residue was transfered to a 250 ml beaker cooled in ice and about 100-110 ml of 50$ aq. acetic acid was used to neutralize the basic solution. The tan precipitate was isolated and washed with a large amount of HgO to remove the acetic acid. After drying in a vacuum oven overnight, 13.07 g (94.32$) of crude 156 product was obtained. The crude solid was purified by consecutive dissolution in aqueous KOH and precipitating by aq. 50% acetic acid. The purified tan solid was Insoluble in most common organic solvents; m.p. 296-298°C (lit. 302-3°C); IR (KBr) 2350-2600(b), 1710, 1280 cm"1; MS, m/e 250. Preparation of Ethvl 2-(2-Pvrldvl)cinchoninate C125J. Method A. The mixture of 2-(2-pyridyl)cinchoninic acid (124),

2.5 g (10 mmol), SOClg (7 ml) and DMF (2 " 3 drops) was refluxed in a 25 ml round-bottomed flask until it became homogeneous. The excess SOClg was removed In vacuo and 8 ml of EtOH was added slowly. The mixture was refluxed for 1 hr, cooled and filtered. Evaporation of the EtOH from the filtrate afforded a crude solid, which was dissolved in a mixture of 10 ml CHCl^ and 5 ml benzene. To this organic solution enough saturated NagCO^ solution was added carefully to neutralize any residual acid. Separation of the organic-phase and removal of CHCl^ and benzene yielded a brown-gray solid, 1.0 g (33.8%); m.p. 71-2°C (lit.213 m.p. 71-2°C); IR (KBr) 1720, 1248 cm"1; 1H NMR (60 MHz,

CDC13) 7.25 - 9.1 (m, 9H), 4.59 (q, 2H, J = 7Hz), 1.51 (t, 3H, J = 7 Hz). MS, m/e = 278. Mfiilmd-JL.213 A mixture of 2-(2-pyridyl)einchoninic acid (124) 157

8.12 g (41 mmol), ethanol 35 ml (0.60 mol) and i^SOy, 5 ml was refluxed for 12 hrs. The solution was cooled in ice bath and neutralized carefully with saturated NagCO^ solution. After neutralization, the suspension was stirred for 30 minutes before the tan preparation was isolated and dried under vacuum. It weighed 7.80 g (86.0?); 7.25 g (80.0?) of purified ester was isolated after recrystallization from HgO/EtOH. Preparation of N-(3-Proovlamino)-2-(2-Pyridyl) cinchoninamide (132). In a 50 ml round-bottomed flask, 1,3-diaminopropane 10 ml (0.12 mol) was heated to reflux before ester 125 3.34 g (13 mmol) was added. The solution turned dark brown while refluxing was continued for 12 hrs. After removal of excess 1.3-diaminopropane by flash evaporation, the residue was treated with HgO, the separated grey powder was filtered off. After drying 1*71 g (41.8?) of crude product was isolated. After recrystallization from benzene, white crystals of 132 1.50 g (36.7?) were collected, m.p. 195-200°C; IR (KBr) 3400, 1630. 1H NMR (60 MHz, DMSO-dg) 7.22-8.75 (m, 9H), 3.56 (q, 2H, J = 6.8 Hz), 2.83 (t, 2H, J = 6.8 Hz), 2.1 (b.s., 3H), 1.76 (t, 2H, J = 6.8 Hz). MS, m/e = 306. 158

Preparation of. Benzvl 2-(2-Pvridvl) cinchoninate (129). 4 g (16 mmol) of 2-(2-Pyridyl)cinchoninic acid (124) was added slowly into a 100 ml beaker containing 1.50 g (0.018 mole) NaHCOg and 40 ml of HgO. The suspension was heated over hot plate and stirred to promote the evolution of CO^ gas. After all the acid was dissolved, the brown solution was filtered and the HgO was evaporated to give tan sodium 2-(2-Pyridyl)cinchoninate (128). The sodium salt was then further dried under vacuum and transferred to a dry 50 ml round bottomed flask, to which a solution of HMPA (30 ml) and benzyl chloride 2.0 g (166 mmol) was added.

The suspension was stirred at room temperature (26-28°C), while monitoring reaction by GLC. After 40 hrs. the clear brown solution was transferred to a 250 ml beaker containing 100 ml of H^O. 60 ml Ethyl ether was used to extract the aqueous mixture three times, the combined ethyl ether layers were dried over sodium sulfate and filtered. After removing ethyl ether, a brown oily material was obtained which was then passed through an alumina column using ethyl ether as the eluent. The purified yellow oil solidified slowly within one week to give 2.50 g (46%) of ester with a banana odor; m.p. 64-65.5°C; IR (KBr) 1720, 1220, 796, 770, 745 cm"1; 1H NMR (200 MHz, CDC13) 7-4-9-1 (m, 14H), 5.51 (3, 2H); MS, m/e 340. Pr_ej>a£ation of 4-Pvridinecarboxvlic Acid (141).216 In a 500 ml three-necked flask fitted with a reflux condenser and stirrer, were placed 5.0 g of 4-picoline (54 mmol), 9-0 g (57 mmol) of potassium permanganate and in 250 ml water; the solution was heated until the purple color had practically disappeared. A second 9.0 g portion of potassium permanganate was then introduced, followed by 50 ml of water, and the heating was continued until the purple color vanished. The reaction was allowed to cool slightly, and the precipiated oxides of manganese were filtered and washed well with 100 ml of hot water. The filtrate was concentrated to 150-200 ml in a beaker, filtered and acidified to pH 3.6 with concentrated hydrochloric acid (65-7.0 ml). The white powder collected after washing with cold water was recrystallized From boiling water, upon cooling in ice, white crystals of acid 141 formed. The yield was 3.18 g (48.1?); m.p. 308-313°C. MS, m/e 123. Preparation of Benzvl 4-Pvridinecar- boxvlate(143).232 In a 50 ml round-bottomed flask was placed 1.26 g (8,7 mmol) of sodium 4-pyridinecarboxylate, 1 ml of p-xylene as glpc internal standard, and 10 ml of HMPA. The reaction mixture was cooled by stirring in an ice 160 bath while benzylchloride 1.10 g (8.7 mmol) was added slowly. After addition the flask was placed in a room temperature water bath and microliter samples were taken at the appropriate time and analyzed by glpc. After 17 hrs. the solution was filtered (sodium salt was all digested) and 45 ml of HpO was mixed into HMPA solution. In a 60 ml separatory funnel, the aq. solution was extracted twice with two 10 ml parts of ethyl ether. The combined ethyl ether layer was dried over MgSO^. A brown liquid was obtained with the characteristic ester odor after removing the ether. The ester was purified by passing through alumina column eluted by ethyl ether to give a colorless liquid which solidified gradually on standing for several days. The crystal leaflets weighed 1.10 g (60SS); m.p. 123-124°C (lit. bp. 2 mmHg, 108-112°C); IR (KBr) 1725, 1592, 750, 700 cm-1; 1H NMR (200 MHz, CDClg) 8.74 (dd, 2H, J = 0.7 Hz, 2.2 Hz, Py-H), 7.83 (dd, 2H, J = 2.2Hz, Py-H), 7.34-7*40 (m, 5H, Ar-H), 5.36 CS, 2H); MS, m/e =

213. Preparation of 4-Benzvloxvpvridine (147).

Method A: (attempted) To 2.5 g (12.4 mmol) of 4-benzyloxypyridine 1-oxide suspended in 30 ml of ice-cooled chloroform, 3 ml of phosphorus trichloride was added and the mixture was heated at 80°C for two hours. After cooling and the 161 addition of water (30 ml), the reaction mixture was made alkaline by 10% sodium hydroxide and extracted with chloroform which was then extracted with HgO. Chloroform extract was dried over magnesium sulfate, and evaporated to dryness. The final brown residue was loaded on a column and eluted with DCM, then ethyl alcohol. After removing DCM and ethyl alcohol, a light-colored viscous liquid was collected in low yield. Method B:

To a 150 ml beaker containing 4.21 g (0.105 mol) of NaOH, dissolved in 4.0 ml of HgO, was added 10 g (0.105 mol) of 4-hydroxypyridine. The red brown solution was heated to evaporate the H^O, and the sodium salt was dried in a vacuum oven overnight. The brown salt mass was broken carefully with a spatula into smaller pieces, into which 40 ml of HMPA was poured and then 12.0 g (94.5 mmol) of benzyl chloride was added dropwise. The resultant mixture was stirred at R.T. for 3 days and then poured into 150 ml HgO in a 500 ml separatory funnel. The aqueous HMPA mixture was extracted three times with 90 ml DCM. After removing DCM the brown residue was purified as in method A and identified by NMR as the same compound. Method C:218 The required sodium salt of 4-hydroxypyridine was prepared as method B, which then was added into a 162 mixture of HgO C30 ml), tetramethyl ammonium chloride (TMAC), 1.15g (10.5 mmol), and toluene (80 ml). To the emulsion 12.0 g of benzyl chloride (94.5 mmol) was added and the resultant mixture was warmed to 80°C for 9 hrs. The aqueous layer was removed and combined with the aqueous wash of the organic layer. The combined aqueous layer was washed with 30 ml DCM. Before a brown viscous mass was collected by evaporating the HgO, the residue was dissolved in hot acetone; after cooling in a refrigerator for 2 days 3.40 g (17.4?) of brown crystals separated. The crude product was passed through an alumina column as described in method A to produce an analytically pure sample; m.p. 55-6°C (hexane) NMR (60 MHz, CDClg) 7.25-7.42 (m, 5H), 6.36-7.561 (dd, 4H, J = 7.3 Hz), 5.08 (s, 2H). MS, m/e = 185. P1 8 Preparation of 4-Hvdroxvpvridine-l-oxide (150). A solution of 12 ml of dimethylaniline in 70 ml of acetic anhydride was heated on the water-bath. To this solution 14 g of 4-nitropyridine-1-oxide was added in portions over a period of 20 minutes. The reaction mixture was heated at 75°C for 20 minutes, then was concentrated in vacuo over 30 minutes at 90°C to remove unreacted AcgO. Addition of 30 ml MeOH to the concentrated solution gave a dark mixture which was cooled in refrigerator overnight. Red-brown crystal were isolated and washed with acetone, dried in vacuo 163

11.30 g (90.2%)of 150 was collected, m.p. 239-240°C (lit. m.p. 239-241° dec.) . Preparation of 3-Benzy_lojcYPvridine. The preparation procedure was the same as that of method B for 4-benzyloxypyridine, except that n-hexane instead of DCM was used to extract the black mass from the resulting organic layer. After purification by column chromatography (diethyl ether, eluent), 36% of the expected product was collected as sticky liquid. 1H NMR (60 MHz, CDClg) 8.1-8.37, 7.1-7.25 (m, 4H, Py), 7.33 (s, 5H) 5.05 (s, 2H). Preparation of 4-Benzvloxvpvridine-1-Oxide IJ149) ,218 To 6.0 g (48 mmol) of 4-nitropyridine-1-oxide in 40 ml of benzyl alcohol was added portionwise a solution of 1 g (0.043 mole) of sodium in 60 ml of benzyl alcohol and the mixture was allowed to stand overnight. The precipitated sodium nitrite was collected and was washed with a little benzyl alcohol. The filtrate and the wash solution were combined, and concentrated in vacuo, where the originally light yellow solution became orange reed. After cooling, the concentrated solution was diluted with ether and allowed to stand. Needles precipitated and were washed with ether and then with acetone, and dried; 6.11 g (63%) of a tan solid was isolated, m.p. 174-176°C (lit 175-177°C); 1H NMR 164

(60 MHz, CDC13) 6.75-8.1 (dd, 4H, J = 7.5 Hz, Py), 6.34 (s, 5H, Ar-H), 5.07 (s, 2H). Preparation of Vinvlbenzvlacetate.

To sodium acetate (2.69 g (33 mmol) in 35 ml of HMPA, 5.0 g (33 mmol) of vinylbenzylchloride was added dropwise and stirred in a 100 ml Erlenmeyer flask at R.T. After 24 hrs, the precipitated white salt was dissolved when "100 ml of HgO was mixed with the HMPA solution. Three 30 ml DCM extracts of aq. HMPA solution were combined and dried over MgSO^. The brown residue after the removal of DCM was chromatographed with diethyl ether as eluent over silica gel; a colorless liquid 4.1 g (71%) was collected. 1H NMR 7.15-7-35 (m, 4H, Ar), 5.1-6.4 (ABX, 3H, H^ = 6.68, Hg = 5.68,

Hx = 5.19, JAB = 17.5 Hz, JAX = 10.5 Hz, JfiX = 1 Hz), 5.03 (s', 2H) 1.99 (s, 3H). IR (neat) 1735, 1610, 1240 cm"1. Attempted Preparation of VinvlbenzvloxvpvridiJie^ 1-oxide. The sodium alkoxide salt of vinylbenzyl acetate was prepared by the following procedure. To a 8.5 ml isopropyl alcohol solution of 3 g (17 mmol) of vinylbenzylacetate, an isopropyl alcohol solution (5 ml) of 0.92 g (14 mmol) of sodium methoxide was injected under Ng, the mixture was stirred at 65°C. After 2.5 hrs, the composition of the reaction mixture was 165 investigated by NMR analysis, the ester had been converted into the corresponding sodium salt. A yellow suspension of 4-nitropyridine 1-oxide (2.38 g, 0.017 mole) in isopropyl alcohol (15 ml) was added. The suspension was stirred at R.T. overnight and was found to remain unreacted. Preparation of Vinvl 4-Benzvloxvpvridlne (155). To 30 ml of aq. NaOH solution was added 5 g (52.6 mmol) of 4-hydroxypyridine to give a red-brown solution. The water was gradually boiled off and the brown salt mass was dried in vacuo. This sodium salt was dissolved in 40 ml of HMPA and then 7.2 g (47.3 mmol) of benzyl chloride was added dropwise. The resultant mixture was stirred at R.T. for 3 days and then poured into 150 ml H20. The aqueous HMPA solution was stirred for 10 minutes and then extracted three times with 90 ml of DCM. After removing the DCM, the residue was purified by column chromatography using ether as eluent; 1.90 g (16?) of semi-solid product was collected. 1H NMR (60 MHz, CDC13) 7.43 (d, 2H, J = 6.5 Hz, Py-H), 7.2 - 7.48 (m, 5H, Ar-H), 6.88 (d, 2H, J = 6.5 Hz, Py-H), 5.2-7.0 (ABX, 3H, H^ = 6.77, Hg = 5.76, Hx = 5.28, JAB = 17.5 Hz, Jfix = 10.5 Hz,

JAX = 1Hz) 5,05 (S' 2H) IR " °» 910 cm-1. Anal. Calc, for Cli(H13N0: C, 79.59; 166

H, 6.20; N, 6.63. Found: C, 79.37; H, 6.18; N, 6.54. MS, m/e = 211. Preparation of 4-Vinvlbenzy-3^2^_(_2-Pvr_idv:Q cinchoniaate (134). To a aqueous NaOH solution prepared from 0-64 g (16 mmol) of NaOH and 30 ml H20, 4 g (16 mmol) of 2-(2-pyridyl)cinchoninic acid (124) was added. The aqueous solution was heated to boil off the H20, the resulting salt was dried in a vacuum, 40 ml of HMPA was introduced, and 2.44 g (16 mmol) of vinylbenzyl chloride was added dropwise. The dark brown solution was stirred at R.T. for 3 days, before it was diluted with 150 ml of HgO. The aqueous HMPA solution was extracted three times with DCM (90 ml) and the combined DCM extract was concentrated in vacuo to give a syrup. Hot n-hexane was used to extract the products; white needles, 2.20 g (38?), separated on cooling. The crystalline material is presumably the p-isomer. A light-colored liquid 1.3 g (22.5?) was recovered from the n-hexene filtrate. The needles were purified by recrystallization from n-hexane, m.p. 82-83.5°C; IR (KBr) 1716, 1586, 990, 900 and 750 cm"1; 1H NMR (60 MHz, CDClg) 7.3-9.1 (m, 13H), 5.24-6.8 (ABX, 3H, HA = 6.74, Hfi = 5.76,

Hx = 5.25; JAB = 17.5 Hz, JAX = 10.5 Hz, Jfix = 1 Hz), 5.5 (S, 2H). Anal. Calc, for C21|HlgN202 167

= C, 78.67; H, 4.95; N, 7.65. Found: C, 78.45; H, 4.93; N, 7.57. MS, m/e = 366. Preparation of Vinvlbenzvl 4-Pvridine- carboxvlate (145). The sodium salt of 6 g (48.7 mmole) of 4-pyridinecarboxylic acid was prepared by neutralization with NaOH as described above. Vinylbenzyl chloride,

7.43 g (48.7 mmole), was dripped into a slurry of sodium 4-pyridinecarboxylate in 40 ml of HMPA, and stirring was continued at R.T. for 3 days. The HMPA solution was mixed with 180 ml HgO and stirred for 15 min.; 100 ml ethyl ether was used to extract the aq. HMPA solution three times. Removal of the ethyl ether yielded a liquid which turned into soft solid upon storage in refrigerator. A sticky semi-crystalline mass 6.60 g (57?) crystallized from hot n-hexane extracts of the ether residue. The crude product can be purified further by column chromatography using ethyl ether as eluent. IR (KBr) 1725, 1605, 1240 cm"1; 1H NMR (60 MHz, CDC13) 7.78-8.81 (2 dd, 4H, J = 4Hz, 1.5 Hz, Py-H), 7.24-7.4 (m, 4H, Ar-H), 5.36 (S, 2H), 5.18-6.96

(ABX, 3H, Ha = 6.74, Hfi = 5.74, Hx = 5.26; JAB = 17.5 Hz, JAX = 10.5 Hz, JBX = 1 Hz). Anal. Calc, for C15H13N02: C, 75.30; H, 5.48; N, 5.85; Found: C, 75.23; H, 5.47; N, 5.74. MS, m/e = 239. 168

The Preparation of Diacefcatopalladium (II), Ed_CQA.cl£ ,233 20 g (6.8 mmol) of NagPdCl^ was reduced with excess aq. hydrazine solution to afford palladium sponge. The black metallic Pd was washed with HgO and dried in vacuo overnight. To a mixture of (18.1 g) HOAc and (0.43 ml) HNO^, dried palladium sponge was added portionwise carefully. Red-brown smoke evolved; thirty minutes later the red brown solution was filtered. The filtrate was subjected to vacuum, HNO^ and HOAc were removed, and orange brown solid separated. After washing with 2-3 ml of conc. HOAc and 2-3 ml of H20, the crystals were dried and weighed; 1.28 g (83.0*), m.p. 184-185°C (dec.), (lit. m.p. 185°C (dec.)). The preparation of tereohthaldehvde (166).23^ A 250 ml three-necked flask was fitted with a 200 ml dropping funnel, and a reflux condenser, the top of which was connected to a gas-absorption trap for HBr, A 300 watt tungsten lamp was mounted within 1 in. of the flask. Into the flask was introduced 50 g (0.48 mol) of p-xylene and the flask was heated in an oil bath maintained between 140°C and 160°C. The stirrer was started and xylene refluxed while 350 g (112 ml, 2.19 mol) of bromine was added gradually through the dropping 169 funnel at such a rate to minimize the unreacted bromine in the flask. Stirring and heating were continued throughout the reaction which requires 3 hrs. After all the bromine had reacted, the mixture was cooled and dissolved in 500 ml of chloroform. The light gray tetrabromo-p-xylene was isolated from the cooled chloroform solution by filtration. In a separate 1 1. round-bottomed flask, the finely powdered crude product was added together with 100 ml. of concentrated sulfuric acid (952). The flask was heated in an oil bath at 70°C and the evolution of red bromine gas was sparged with air. When the evolution of gas becomes less vigorous the temperature was raised to 100°. After the evolution of bromine ceased, the mixture was poured into 300 g of crushed ice. The crystalline solid (8.2 g (72.8?)) was collected, m.p. 113-115°C, (lit. m.p. 115-116°C). Preparation of p-Bromobenzvlalcohol (172). A solution of p-bromobenzaldehyde 5 g (27 mmol.) in ethanol 40 ml was added to a well-stirred ethanol solution (40 ml) of excess NaBH^ (0.357 g). The reaction mixture was stirred for 1 hr. at 25°C in a water bath then another hr. without water bath. Ether (80 ml) was employed to extract the mixture twice. The combined extract was washed with HgO, after drying over MgSOjj, ether was removed to produce 4.23 g 170

(83.8%) of white product, m.p. 75-76°C (lit.2^ m.p. 78.5-79°C) IR (KBr)) 3450 (-0H), 1H NMR (60 MHz,

CDC13), s 7.32-7.78 (d of q, 4H, J = 8.5 Hz, 2 Hz, Ar-H), 4.77 (s, 2H, - CHgO), 3.72 (s, 1H, -OH). MS, m/e = 186. Attempted Synthesis of.4-VinvIbenzaldehvde (170). A 250 ml three-necked round-bottomed flask was fitted with a nitrogen inlet and a gentle flow of nitrogen through the apparatus was maintained throughout the reaction. An ethereal solution of n-butyl lithium (25 mmol, about 12 ml of 2.1 M n-Buli solution) and 50 ml of anhydrous ether was added to the flask. The solution was stirred and 8.93 g (25 mmol) of triphenylmethyl-phosphonium bromide was added cautiously over 5-minute period. The solution turned yellow-orange, stirring was continued at room temperature for 4 hrs. The solution of phosphorane was transfered to a septum-capped separatory funnel, and added dropwise to a

suspension of terephthaldehyde 3.35 g (25 mmol) suspended in 70 ml of ether. The solution was heated to reflux overnight under nitrogen. The milky suspension was quenched with 30 ml HgO and filtered to give the ether layer which was concentrated to give a light yellow sticky mass which showed no sign of alkene formation by NMR. 171

£r-epar.at.i.oiL-Q.f-D.3,bensalacgtQne (DBA). A cooled solution of 10 g of NaOH in 100 ml of water and 80 ml of ethyl alcohol was placed in a 500 ml wide-mouthed glass jar which was surrounded with HgO and fitted with a magnetic stirrer. The solution was kept at about 20-25°C and stirred vigorously while one-half of a mixture of 10.6 g (0.1 mole) of benzaldehyde and 2.9 g (50 mmol) of acetone was added. A yellow cloud formed which soon became a flocculent precipitate. After fifteen minutes the rest of the mixed reagents was added. Vigorous stirring was continued for one-half hour and the mush was then filtered with suction on a Buchner funnel, washed with distilled water and dried in vacuo, 11.08 g (9*1.7?) of pale yellow dibenzalacetone was isolated m.p. 107-109°C (lit. m.p. 110-111°C).

Prepara t.ioiL-Q-f _fcJie, _£fl^ (DPAJ.^C >2 3 ^ NagPdCljj (0.87 g (2.96 mmol)) was added to hot (ca. 50°C) methanol (75 ml) containing DBA 2.30 g (9.8 mmol) and sodium acetate 1.95 g (23.8 mmol). The mixture was stirred for *1.5 hrs at 40°C to give a reddish-purple precipitate and allowed to cool to complete the precipitation, the precipitate was removed by filtration, washed successively with water and acetone and dried in .vacuo.. The precipitate was dissolved in hot chloroform (60 ml) and filtered to give 172 a deep violet solution. To this solution, diethyl ether 85 ml was added slowly. Deep purple needles precipitated. They were removed by filtration, washed with diethyl ether, and dried in vacuo. The complex, m.p. 120-123°C dec. (lit. 122-124°C dec), was obtained in 75? yield. Preparation of 15t DVB Cross-linked I Poly(vinylp^ridine)1 (I).230 PolyCvinyl alcohol), 0.48 g, was dissolved in 110 ml of boiling water and the solution was placed in a 500 ml resin kettle equipped with a sealed ground glass stirrer bearing, a reflux condenser, a nitrogen inlet and a mechanical stirrer. The solution was stirred under nitrogen at 80°C and a solution of 5.0 g of 4-vinylpyridine and 0.75 g of divinylbenzene in 10 ml of toluene was added rapidly. After addition of 0.2 g of azobisisobutyronitrile (AIBN), the polymerization was allowed to proceed under constant vigorous stirring. Polymer beads started to appear very rapidly but the mixture was left at 80°C overnight. The resin beads were collected by filtration and were washed extensively with water, ethanol, ether, CHCl^, and finally methanol. After drying in vacuo for two days, 5.73 g of polymer was obtained. Elemental analysis indicates the presence of 11 .2116 nitrogen, equivalent to 8.01 173

mequiv. of pyridine per gram of polymer. Anal. Found: C, 77.28; H, 6.55; N, 11.21. Preparation of DVB Cross-linked iPolv(4-vinvlbenzvloxvpvririine)1 (II) The set-up and procedure are the same as those for the preparation of DVB cross-linked poly(vinylpyridine)] (I), except that the reaction was proceeded with 4-vinylbenzyloxypyridine 11.2 g (5.34 mmol), technical grade DVB 0.12 g, AIBN 0.05 g, polyvinylalcohol 0.12 g, toluene 10 ml and distilled and degassed (freshly boiled) H20 (50 ml). White polymer resin beads 1.10 g (88.7?) which contained 3.8 mequiv. per gram pyridine were isolated. Anal. Found: C, 78.75; H, 6.53; N, 5 .82. Pr_e.para.tion of DVB Cross-linked (Polv(vinvlbenzvl-4-pyr_idinecarboxvlate) 1 (III).

The procedure is the same as that for the preparation of DVB cross-linked DVP (I), except that the reaction was charged with vinylbenzyl 4-pyridinecarboxylate, 6.0 g (26.4 mmol), DVB 0.3 g f AIBN 0.2 g, poly(vinylalcohol), 0.6 g, 100 ml of degassed HgO and toluene 10 ml. 6.0 g (95.256) of polymer resin beads were isolated. The nitrogen content of 5-3056 is corresponding to 3.8 mequiv. of pyridine unit per gram of polymer. IR (KBr) 1730 cm“^. 174

Anal-. Found: C, 75.4; H, 5.73; N, 5.30. Preparation of DVB Cross-linked fPolv(vinvlbenzvl 2-(2-Pvridvl)cinchoninate1 CIV). The reagents used were vinylbenzyl 2-(2-pyridyl) cinchoninate (2.50 g, 6.8 mmol), divinylbenzene, 0.125 g, AIBN, 0.06 g, poly(vinylalcohol), 0.2 g, toluene, 6 ml and distilled H20, 60 ml; 2.4. g (91.4?) of polymer was isolated. Based upon elemental analysis, there are 2.2 mequiv. of dipyridyl unites per gram of polymer. IR (KBr) 1736 (-C02R) cm”'*. Anal. Found: C, 76.76;

H, 5.43; N , 6.15. Preparation of 15? DVB Cross-linked F Poly.C vinyl pyridine.) PdCl£ 1 (VI).238 To a solution of 0.120 g of Na2PdCl^ in 30 ml of absolute ethanol in a 50 ml Erlenmeyer flask was added 30 ml DCM 1.3 g of swollen 15? DVB cross-linked poly(vinylpyridine (I). The suspension was stirred at room tempperature for 1 day, filtered and washed with ethanol. The yellow polymer supported catalyst was dried. The filtrate was subjected to colorimetric analysis for palladium, almost no Na2PdCl^ in ethanol solution (< 10“^ grams) was detected. The palladium complex of 10? DVB cross-linked resins was also prepared. 175

Er-eparation of rFolv(vinvlbenzvloxv-4-pvridine) PdCl£J (VIII). Follow the same method above, by the treatment of DVB cross-linked poly (vinylbenzyloxy-4-pyridine), 0.811 g (3.08 mequiv.) with NagPdCl^ 0.0294 g (0.1 mmol) yielded 0.82 g of light-yellow polymer. The filtrate was checked by U.V.-vis spectrophotometer, a negligible amount of Pd was found (< 10"® g). Preparation of DVB Cross-linked TPolv(vinvlbenzvl 4-pvridinec.arboxvlate) PdCl„] (VII). Follow the same procedure as above by the treatment of polymer III, 1.20 g (3.08 mequiv. of pyridine units) and NagPdCl^, 0.0294 g (0.1 mmol). The desired polymeric complex was formed quantitatively. Ge_nera 1__Viny_lic__Snb.sbl_tution.Procedure for Homogeneous Catalysis. The aryl halide (10 mmol), olefin (10 mmol), 0.1 mmol of sodium tetrachloropalladate, 0.3 ml (3.08 mmol) of 4-picoline and 1.04 g of p-xylene were combined in a capped 150 x 20 mm test tube. If aryl iodide was used, 11 mmol of morphollne and 3.5 ml of DMF were added. If the substrate was an aryl bromide, 12 mmol of NaHCO^ and 3*5 ml HMPA were added. The tube was immersed in an oil bath at 90°C and stirred with a magnetic stirring bar. The reaction was monitored by periodic G.C. 176 analysis, the decrease of olefin or aryl halide was followed. The products were usually recovered, after 85$ completion of reaction, by adding water and filtering the insoluble solid products, or by adding ether and water and extracting the product. The product extracts or solids were washed with aqueous acid to remove excess amine or sodium bicarbonate. The crude product were then purified by recrystallization from ethanol or other suitable solvents. 4-Picoliine may be replaced by 3.08 mmol of a variety of substituted pyridines or dipyridyls. General._Vinvlic Substitution Procedure_for Polymer-Suppor-ted Catalysis... The procedure was the same as the previous one except 4-picoline was replaced by a polymer-supported catalyst. The catalyst could be isolated by vacuum filtration and washing with small amount of DMF. The product was isolated from the filtrate by diluting with HgO. The water dissolved the HMPA and allowed the stillbene derivatives to be taken up in ether. Evaporation of the ether and recrystallization from ethanol-HgO mixtures appeared the pure products. (El-H-Methoxvstiibene^ White solid; m.p. 135-136°C (lit. m.p. 136°C);226,228 IR(KBr) l6o4, 1510, 1258, 976, 825,

760 cm"1; 1H NMR (200 MHz, CDClg), 7.20-7.50 (m, 177

7H), 7.08 Cd, 1H , J = 8.8 Hz, olefinic H), 6.95 (d, 1H, J = 8.8 Hz, olefinic H), 6.9 Cd, 2H, Ar-H), 3.65

(E)-4-Methylstil.bene., White solid; m.p. 119-120°C (lit. m.p. 119.2-119.8°°C);226 IR(KBr) 1588, 970, 810, 750 cm"1; 1H NMR (60 MHz, CDC13). 7.13-7.4 (m 9H, Ar-H), 7.01 (s, 2H, olefinic), 2.34 (s, 3H, -CHg). MS, m/e = 194.

(E)rStilhene-«. White solid; m.p. 122-124°C (lit. m.p. 124°C);226’228 IR(KBr); 1H NMR (60 MHz, CDC13), 7.15-7.50 (m, 10H, Ar-H), 7.07 (s, 2H, olefinic H). MS, m/e = 180. ( E)-4-Stilbeneamlne.. Purple brown solid; m.p. 140-143°C (lit. 144°C); IR(KBr) 3430-3330 (b), 1540, 965, 820, 755 cm"1, 1H NMR (200 MHz), CDClg), 7.2-7.5 (m, 7H, Ar-H), 7.03 (d, 1H, J = 8.1 Hz, olefinic H), 6.89 (d, 1H, J = 8.1 Hz, olefinic H), 6.65 (d, 2H, Ar-H), 1.57 178

(broad s, 2H, -MHz). MS, m/e = 195. (E)-4-TrifluoromethvlstiIbene. Solid; m.p. 133-134°C (lit. m.p. 132.6-133.6°C);226 IR(KBr) 1615, 1333, 978, 750 cm"1; 1H NMR (60 MHz, CDClg) 7.2-7.5 (m, 9H, ArH), 7.07 (s, 2H, olefinic). MS, m/e = 248. (El-4-Chlorostilbene. Solid, m.p. 127-128°C (lit. 128°C);225'228 IR(KBr) 1595, 984, 835 cm"1; 1H NMR (60 MHz, CDC13). 7.2-7.43 (m, 9H, Ar-H), 6.98 (s, 2H, olefinic H). MS, m/e = 214. M-Fluorostilbene. Solid; m.p. 64-66°C (lit. 73°C); IR(KBr) 1616, 786. 758, 700 cm"1; 1H NMR (200 MHz, CDClg) 7.13-7.35 (m, 9H, Ar-H), 7.0 (S, 2H, olefinic). MS, m/e

= 198. Bibliography

1. T. Mizoroki, K. Mori and A. Ozaki,

flails— Chem. Soc. J pn», 1973, 46, 1505. 2. R.F. Heck and J.P. Nolley, Jr., J_._Qrg. Chem.t

1972, 37, 2320. 3. A.J. Chalk and S.A. Magnennis, J_._0rg.. ChenL.,

1976, 41, 273. 4. H.A. Dieck and R.F. Heck, J. Am. Chem. Soc.f

1974, 96, 1133. 5. J. Smidt, V. Hafner, R. Jira, R. Sieber, J. Sedlmier and A. Sabel. Angew._ _ChenL. 74, 93 (1962); Angew. Chem. Intern. Ed.r 1962, 1, 80. 6. P.M. Henry, Adv. Organometal. Chem^. 1975, 13,

363. 7. E.W. Stern, M.L. Spector and H.P. Leftin,

J , CataLi, 1966, 6, 152. 8. R.D. Rieke, W.J. Wolf, N. Kujundzic and A.V. Kavaliunas, J■^_AmJ_Chem. Soc.f 1977, 99, 4159. 9. D .R. Coulson, Inorg. Svnth.. 1972, 13, 121. 10. K. Blttler, N.V. Kutepow, K. Neubauer and H. Reis, Angew. Chem.f 1968, 80, 352. 11. P. Roffia, F. Conti and G. Gregorio, Chim. Ind. (Milanlf 1972, 54, 317.

179 180

12. L. Malatesta and M. Angoletta, J . Chem. Soc. r 1957, 1186.

13. P. Roffia, G. Gregorio, F. Conti, G.F. Pregaglia and R. Ugo, J. Mol. Catal., 1977, 2, 191. 14. E.V. Greaves, C.J.L. Lock and P.M. Maitlis, Can. J. Chem.. 1968, 46, 3879.

15. G. Wilke, E.W. Muller, M. Kroner, P. Heimbaeh and H. Breil, Fr. Pat. 1 320 729 (1963) (Chem. Abstr., 1963, 59, 14026). 16. B.L. Shaw and J. Powell, J. Chem. Soc. (A), 1968, 774.

17. W. Kuran and A. Musco, J. Orcanomet. Chem.. 1972, 40, C47. 18. a) B.M. Trost, T.A. Runge and L.N. Jungheim, J. Am. Chem. Soc.r 1980, 102, 2840. b) P.S. Elmes and W.R. Jackson, J . Am. Chem.. _Soc^f 1979, 101, 6128. c) Y. Inoue, T. Hibi, M. Satake and H. Hashimoto, J. Chem. Soc.r Chem. Commun..

1979, 982.

19. Y. Takashi, T. Ito, S. Sakai and y. Ishii, J. Chem. Soc.f.Chem. Commun.. 1980, 1065. 2 0 . T. Ukai, H. Kawazura and Y. Ishii, J. Organomet. Chem.. f 1974, 65, 253-266.

21 . A.O. King, E. Negishi, F.J. Villani, Jr. and A. Solveira, Jr., J. Ore. Chem.. 1978, 43, 358. 161

22. K. Kudo, M. Hidai and Y. Uchida, J.. Organomet.

Chsnu, 33, 393 (1971). 23. M. Hidai, H. Kokura and Y. Uchida, J. Organomet.

Ch.em^f 52, 431 <1973). 24. J.H. Darling and J.S. Ogden, J. Chem. Soc.,

Dalton Trans.. 1079 (1973). 25. E.P. Kundig, M. Moskovits and G.A. Ozin, Can .-J-. ChenL.. 50, 3587 (1972). 26. L. Malatesta, J. Chem. Soc.T 3924 (1955). 27. L. Malatesta, Reel. Trav. Chim. Pavs-Bas. 75,- 644 (1956). 28. S. Otsuka, T. Yoshida, M. Matsumoto and K. Nakatsu, J . Am .-.Chem. Soc.f 1976, 98, 5850.

29. P. Fitton and J.E. McKeon, Chem. Commun., 1968, 4. 30. J .A. Osborn, Chem. Commun.. 1968, 1231. 31. B.E. Mann and A. Musco, J. Chem. Soc., Dalton

Inaufi*, 1975, 1673. 32. H. Werner and W. Bertleff, J. Chem. Res. (S),

1978, 201. 33. S. Otsuka, A. Nakamura, T. Yoshida, M. Naruto and K. Ataka, J.«_Am_.—Chem. Soc., 1973, 95, 3180. 34. G.W. Parshall, J. Am. Chem.. Soc.T 1974, 96, 2360. 35. P. Fitton and E.A. Rick, J. Organomet. Chem..

1971, 28, 287. 182

36. A.V. Kramer, J.A. Labinger, J.S. Bradley and J.A. Osborn, J. Am. Chem. Soc., 1974, 96, 7145; A.V. Kramer and J.A. Osborn, J. Am. Chem. Soe.T 1974, 96, 7832. 37. J.J. Levlson and S.D. Robinson, Chem. Commun.,

1967, 198. 38. a) A.J. Mukhedkar, M. Green and F.G.A. Stone, J. Am. Chem. Soe. (A)r 1969, 3023. b) H.D. Empsall, H. Green, S.K.S. Shakshook1 and F.G.A. Stone, J. Chem. Soc.. (A)f 1971, 3472. 39. a) J. Burgess, M.E. Howden, R.D. Kemmltt and N.S. Srldhara, J. Chem. Soo.r Dalton Trans., 1978, 81. b) G.A. Chukhadzhayan, Z.K. Evoyan and L.N. Melkonya, Zn. Obshch. Khim.f 1975, 45, 1114. 40. a) B. Croclani, M. Nicolini, D.A. Clemente and G. Bandoli, J. Organomet. Chem.r 1973, 49, 249. b) V.I. Sokolov, V.V. Bashilov, L.M. Anishchenko and 0.A. Rentov, J. Organomet. ChenL., 1974, 71, C41. c) V.I. Sokolov, V.V. Bashilov, O.A. Rentov, M.N. Bochkarev, L.P. Mayorova and G.A. Razuvaev, J. Organomet.- Chem. T 1976, 112, C47. d) V.T. Bychkov, S.N. Titova and I.V. Lomakova,

Zh. Obshcb^-KhinL, 1976, 46, 2629 (Ctem., Abatis, 1977, 86, 72852). 41. Y. Takahashi, T. Ito, S. Sakai and Y. Ishii, 183

Chem ■ Commun.. 1970, 1065. 42. Y. Ishii, S. Hasegawa, S. Kimura and K. Itoh, Jt Organomet. Chem., 1974, 73, 411. 43. a) T. Ito, Y. Takahashi and Y. Ishii, J. Chem. _Soc.■ Chem. Commun., 1972, 629. b) T. Ito, S. Hasegawa, Y. Takahashi and Y. Ishii, ibid . T 1972, 629. 44. T. Ito, S. Hasegawa, Y. Takahashi and Y. Ishii,

J» OrganomeJL._C,hea.., 1974, 73, 401 . 45. a) G. Allegra, A. Immirzi and L. Porri, J. Am. Chem. Soc■, 1965, 87, 1394. b) G. Allegro, G.T. Casagrande, a. Immirzi, L. Porri and G. Vitulli, I h l i 1970, 92, 289. 46. Y. Kobayshi, Y. Iitaka and H. Yamazaki, Acta Crvstallogr.T 1972, B28, 899. 47. a) A. Ducrulx, H. Felkin, C. Pascard and G.K. Turner, J. C h e m .... Soc. r Chem. Commun.. 1975, 615. b) H. Felkin and G.K. Turner, J._Organomet. Chem.f 1977, 129, 429. c) A. Ducruix and C. Pascard, Acta Crvstallogr .f 1977,

£31, 3688. 48. S. Otuka, Y. Tatsuno, M. Hiki, T. Aoki, M. Hatsumoto, H. Yoshioka and K. Nakatsu,

jL._Chem.. Soc.. . C h e m C o m m u n . f 1973, 445.

49. J. Dubrawski, J.C. Kriege-Simondsen and R.D. Feltham, J ..Am ...Chem. Soc.. t 1980, 102, 2089. 184

50. A.F. Wells, Z. Krlst, 1938, 100, 189. 51. H. Shafer, V. Wiese, K. Rinke and K. Brendel, Angew.__Chem. Int. Ed. f 1967, 6, 253. 52. J. Tsuji, "Organic Synthesis with Palladium Compounds", Springer-Verlag, Berlin, 1980. 53. R. Van Helden and T. Jonkhoff (Shell Oil Co.), U.S.P. 3,210,152 (1965); C.A. 1965, 63, 15895a. 54. a) M.S. Kharasch, R.C. Seyler and F.R. Mayo, J. Am. Chem. Soc,f 1938, 60, 882. b) J.R. Doyle, P.E. Slade and H.B. Johassen,

Inorg.-JSynttL., I960, 6, 216. c) G. Piaro, N. Netto, A. Musco and R. Palumbo, Ric. Sic. Parte 2,

Sez, 1965, A8, 1441. 55. J.W. Mellor, A Comprehensive Treatise on Inorganic and Theoretical Chemistry, Longmans, Green, New York (1936). 56. J.R. Durig, R. Layton, D.W. Sink and B.R. Mitchell, Spectroohlm. Acta.f 1965, 21, 1367. 57. T.H. Black, Aldrichlmica Acta.. 1982, 15, 13.

58. R.F. Heck, Pure and Ad d I . Chem.. 1978, 50,

691. 59. R.F. Heck, J. Am. Chem. Soc.f 1968, 90, 5518. 60. J.E. Plevyak and R.F. Heck, J. Org. Chem., 1978,

43, 2454. 61. C.B. Ziegler, Jr., and R.F. Heck, J. Org. Chem..

1978, 43, 2941. 185

62. a) J.B. Melpolder and R.F. Heck, J . Org. Chem., 1976, 41, 265. b) A.J. Clark and S.A. Magennls, J. Qrg. Chem..., 1976, 41, 273- 63. a) M. Mori, K. Chiba and Y. Ban, Tetrahedron

LfitiLa., 1977, 1037. b) M. Mori and Y. Ban, Tetrahedron Lett., 1976, 1803, 1807; 1979, 1133; M. Mori and Y. Ban, Hetfir-O.cyglfia, 1978, 9, 391. 64. R. Odle, B. Blevins, M. Ratcliff and L.S. Hegedus, J. Org. Chem.. 1980, 45, 2709. 65. H. Ida, Y. Yuasa and C. Kibayashi,

J.fc-0rg-i-.Chem», 1980, 45, 2938. 66. F. Ziegler, V. Chakrabortz and R.B. Weisenfeld, Tetrahedronf 1981, 37, 4035. 67. a) R.C. Larock and M.A. Mitchell, J. Am. Chem. Soc., 1978, 100, 180. b) R.C. Larock, J.C. Bernhardt and R.J. Driggs, J. Organomet. Chem.f 1978, 156, 45. 68. B.A. Patel and R.F. Heck, J. Qrg. Chem., 1978, 43, 3898. 69. a) K. Kaneda, F. Kawamoto, Y. Fujiwara, T. Imanaka and S. Teranishi, Tetrahedron Lett., 1974, 1067. b) K. Kaneda, T. Uchiyama, Y. Fujiwara, T. Imanaka and S. Teranishi, J. Org. Chem.T 1979, 44, 55. 70. H.A. Dieck and R.F. Heck, J. Qrg. Chem., 1975, 40, 1083. 186

71. D.D. Gender, L.C, Kao, J.I. Kim, 6.A. Patel and R.F. Heck, University of Delaware, unpublished results. 72. B.A. Patel, J.E. Dickerson and R.F. Heck, J. 0rg_.__Chem.. 1978, 43, 5018.

73. B.A. Patel, L.C. Kao, N.A. Cortese, J.V. Minkiewicz and R.F. Heck, J. Org. Chem.f 1979, 44, 918.

74. B.A. Patel and R.F. Heck, J. Ore. Chem.T 1978, 43, 3898.

75. W. Fischetti, K.T. Mak, F.G. Stakem and R.F. Heck, J_. Org. .ChenL.. 1983, 48, 948. 76. J. Smidt, et al., Angew. Chem..f 1959, 71 , 176.

77. I.I. Moiseev, M.N. Vargaftic and Ya.K. Syikin, Dokl. Akad. SSR. 1960, 133, 377. 78. E.W. Stern and M.L. Spector, Proc. Chem. Soc., 1961, 370.

79. J. Tsujl, Pure In. Ap d I. Chem.. 1981, 53, 2371. 80. T. Hosokawa, H. Ohkata and I. Moritani, Bull. Chem. Soc. Jpn.T 1975, 48, 1533. 81. T. Hosokawa, S. Yamashita, S-i Murahashi and A. Sonoda, Bull. Chem. Soc. Jpn.. 1976, 49, 3662.

82. T. Hosokawa, S. Miyagi, S-i Murahashi and A. Sonoda, J. Org. Chem.T 1978, 43, 2752.

83. T. Hosokawa, T. Uno, S. Inui and S-i Murahashi, 187

J. Am. Chem. Soc.,. 1981 , 103, 2318. 84. T. Izumi and A. Kasahara, Bull. Chem. Soc. Jpn. T 1975, 48, 1673. 85. B. Akermark, J.E. Backvall, K. Siirala-Hansen, K. Sjoberg, K. Zetterberg and L.S. Hegedus, J. Qrganoroet. .Chem,,., 1974, 72, 127. 86. L.S. Hegedus, B. Akermark and K. Zetterberg, unpublished results. 87. J.E. Backvall and E.E. Bjorkman, J. Org. Chem., 1980, 45, 2893. 88. L.S. Hegedus, EC. Siirala-Hansen, J. Am. Chem. Soc.. 1975, 97, 1184. 89. L.S. Hegedus, O.P. Anderson, K. Zetterberg, G. Allen, K. Siirala-Hansen, D.J. Otsku and A.B. Packard, Inorg ,_ Chem ., 1977, 16, 1887. 90. L.S. Hegedus, G.F. Allen, J.J. Bozell and E.L. Waterman, J . _Am_.__Chem. Soc . T 1978, 100, 5800. 91. L.S. Hegedus and J.M. Mckearin, J. Am. Chem. Soc.. 1982, 104, 2444. 92. L.S. Hegedus, P.M. Winton and S. Varaprath, J. Ore. Chem.. 1981, 46, 2215. 93. S. Hatano, M. Saruvatare, EC. Isomura and H. Taniguchi, Heterocvcles. 1981, 15, 747. 94. A. Kasahara, T. Izumi and 0. Saito, Chem. Ind.t (London), 1980, 666. 95. D.E. Korte, L.S. Hegedus and R.K. Wirth, 188

,J-«-p£g-,_Chem.,, 1977, 42, 1329. 96. A. Kasahara and N. Fukuda, Chem. Ind., (London), 1976, 485.

97. S-i. Murahashi, M. Yamamura and N. Mita, J. Ore. Chem.f 1977, 42, 2870. 98. L.S. Hegedus, R.E. Williams, M.A. McQuire and T. Hayashi, J_^_Am. Chem. Soc.. 1980, 102, 4973.

99. B.M. Trost, Tetrahedron Lett.. 1977, 33, 2615. 100. J.K. Stille and R. Divakaruni,

J. Oreanomet. Chem.T 1979, 169, 239. 101. J.E. Backvall, B. Akermark and S.O. Ljunggren, J. Am. Chem. Soo.t 1979, 101, 2411. 102. a) P.M. Henry, G.A. Ward, J. Am. Chem. Soc.f 1971, 93, 1494. b) D.S. Andell, J.E. Backvall, J. Organomet. Chem.. 1983, 244, 401.

103. D.E. James, L.F. Hines and J.K. Stille, J. Am. Chem. Soc.r 1976, 98, 1806.

104. H. Kurosawa, T. Majima and N. Asada, J. Am. Chem. Soc.. 1980, 102, 6996.

105. B. Akermark, J.E. Backvall, K. Siiraralla-Hansen, K. Sjoberg and K. Zetterberg, Tetrahedron Lett.. f 1974, 1363. 106. H. Kurosawa and N. Asada, Tetrahedron Lett., 1979, 255. 107. D.M. Henry and G.A. Ward, J. Am. Chem. Soo.T 1972, 94, 673. 189

108. A. Seqnitz, D.M. Bailey and P.M. Maitlis, J_._ Chem Soc .. Chem. Commun . f 1973, 698. 109. S.I. Murahashi, M. Yamamura and N. Mita, J. Ore. Chem.. 1977, 42, 2870. 110. a) K. Fukui, Aoc. Chem. Res.. 1971, 4, 57. b) K. Fukui, J. Chem. Phvs.. 1957, 27, 1247. 111. G. Klopman, J._,_Jhn._Chem. Soc.T 1968, 2, 90. 112. G. Klopman, Ed., "Chemical Reactivity and Reaction Paths", Wiley, New York, 1974. 113. C.N. Satterfield. "Heterogeneous Catalysis in Practice"; McGraw-Hill: New York, 1980. 114. J. Falbi and H. Bahrmann, J.__Chem. Educ.., 1984, 61, 961. 115. S.C. Stinson, C&EN. 1983, 19. 116. J. Manassen, Platinum Met. Rev.f 1971, 15, 142. 117. D.E. Bergbreiter, B. Chen and T.J. Lynch, J. Org^Chem^. 1983, 48, 4179. 118. R. Linarte-Lazcano, M.P. Pedrosa, J. Sabadie and J.E. Germain, Bull. Soc .- Chim ., 1974, 1129. 119. W.0. Haay and D.D. Whitehurst (Mobile Oil Co.), Ger. Pat. 1 800 380. 120. J.F. McQuillin, W.O. Ord and P.L. Simpson, J. Chem. -Soc. 1963, 5996. 121. G.G. Allan and A.N. Neogi, J. Catal., 1970,

19, 256. 190

122. S.C. Tang, T.E. Paxson and L. Kim, J. Mol. CataL... 1980, 9, 313. See also U.S. Pat. M 179 if02 and U.S. Pat. if 179 M03- 123. J.P.J. Verlaan, J.P.C. Bootsma, C.E. Koning and G. Challa, J. Mol. Catal., 1983, 18, 159. 12if. Y.J. Li and Y.Y. Jiang, J. Hoi. Catal.f 1983,

19, 277. 125. Y.J. Li and Y.Y. Jiang, J. Mol. Catal.. 1983,

19, 283. 126. G. Challa, J. Mol. Catal., 1983, 21, 1. 127. W. Heitz, Adv. Polvm. Sol.. 1977, 23, 1. 128. K.W. Pepper, H.M. Paisley and M.A. Young,

J. Chem. Soc.f 1953, 4097. 129. a) C.U. Pittman and G.O. Evans, Chem. Teclmpl.r 1973, 560. b) C.U. Pittman and R.F. Felis, J. Organomet. Chem.. 197if, 72, 389. 130. R.H. Grubbs, C. Gibbons, L.C. Kroll, W.D. Bonds and H. Brubaker, J. Am. Chem. Soc.r 1973, 95,

2373. 131. M. Kraus and D. Tomanova, J. Polvm. Sol. Polvm. Chem. Ed.. 197M, 12, 1781. 132. R.H. Grubbs and L.C. Kroll, J. Am. Chem. Soc.r

1971, 93, 3062. 133. N.L. Holy, J. Org. Chem.r 1979, 44, 385. 13if. W.O. Haag and D.D. Whitehurst in: Catalysis, 1, Paper 29, M65, J.W. Hightower, Ed., 1973. 191

135. R.S. Drago and J.H. Gaul, J. Chem. Soc.f Chem. Commun.r 1979, 746. 136. R.H. Grubbs, C. Gibbons, L.C. Kroll, W.D. Bonds and H.C. Brubaker, Jr., J. Am. Chem. Soc., 1973,

95, 2373. 137. B.H. Chang, R.H. Grubbs and H.C. Brubaker, Jr., J. Orcanomet. Chem., 1979, 172, 81. 138. K.G. Allum and R.D. Hancock (British Petroleum), Br. Pat. 1 277 737 (1972) and 1 295 673 (1972).

139. M. Capka, P. Svoboda, M. Corny and J. Hetflejs, Tetrahedron .Lett,.., 1971, 4787. 140. R.H. Grubbs and L.C. Kroll, J. Am. Chem. Soc.,

1971, 93, 3062. 141. R.H. Grubbs and E.M. Sweet, Macromo_leculesr

1975, 8, 241. 142. J. Kiji, S. Kadoi and J. Furukawa, Ancew. Markromol. Chem.. 1975, 46, 163.

143. C.U. Pittman, Jr. and L.R. Smith, J. Am. Chem. Soc.. 1975, 97, 1949. 144. C.U. Pittman, Jr., L.R. Smith and R.M. Hanes, J. Am. Chem. Soc_.. 1975, 97, 1742.

145. C.U. Pittman, Jr., S.E. Jacobson and H. Hiramoto, J. Am. Chem. Soc., 1975, 97, 4774. 146. J.M. Basset, R. Mutin, G. Descotes and D. Slnou, C.R. Acad. SclSer.r C, 1975, 280, 1181.

147. G.O. Evans, C.U. Pittman, Jr., R. McMillan, R.T. 192

Beach and R. Jones, J. Orcanomet. Chem.r 1974,

67, 295. 148. N. Haglhara, S. Takahashi and Y. Nonaka (Osaka U.), Jap. Pat. 4 813 487 (Chem^Abstr., 1973, 78, 148481). 149. L.D. Rollman, Inorc. Chlm. Acta.. 1972, 6,

137. 150. J.P. Collman, L.S. Hegedus, M.D. Cooke, J.R. Norton, G. Dolcetti and D.N. Harquardt, J. Am. Chem. Soc.T 1972, 94, 1789. 151. K.G. Allum and R.D. Hancock, Preprint 31, 5th International Congress of Catalysis, Miami, 1972. 152. British Petroleum, British Patent 2647475 (Invs. K.G. Allum, R.D. Hancock and M.J. Lawrenson). 153. M. Capka, P. Svoboda, M. Cerny and J. Hetflejs, Tetrahedron Lett.f 1971, 4787. 154. G.O. Evans, C.U. Pittman, r. McMillan, R.T. Beach and R, Jones, J. Oreanomet. Chem.. 1974, 67,

295. 155. R.H. Grubbs and L.C. Kroll, J. Am. Chem. Soc..

1971, 93, 3062. 156. F. Hojabri, Polymer, 1976, 17, 58.

157. F. Hohabrl, J. A p p I. Oiem^^Blotechnol ■ r 1976, 26, 382. 158. C.C. Leznoff, Chem. Soc. Rev.r 1974, 3, 65. 159. K .A. Abdula, N.P. Allen, A.H. Badran, J. Dwyer, 193

C.A. Mcauliffe and N.D.A. Toma, Chem. Ins., 1976, 273.

160. K.G. Allum, R.D. Hancock, 2.V. Howell, R.C. Pitkethly and P.J. Robinson, J. Orcanometal. Chem.r 1975, 87, 189. 161. British Petroleum, British Patent 1295673 CInvs. K.G. Allum and R.D. Hancock). 162. British Petroleum, British Patent 1291237 CInvs. K.G. Allum and R.D. Hancock).

163. C.U. Pittman, Jr. and C.C. Lin, J. Org. Chem., 1978, 43, 4928.

164. C.U. Pittman, Jr. and A. Hiraro, J. Org. .ChfiPU, 1978, 43, 640.

165. R.D. Sanner, R.G. Austin, M.S. Wrighton, V.D. Honnick and C.U. Pittman, Jr., Adv. Chem.. SetL. f 1980, 184, 13. 166. V. Dumont, J.C. Paulin, T.P. Dang and H.B. Kagan,

Chem. Soc,., 1973, 95, 8295.

167. J.C. Paulin, W. Dumont, T.P. Dand and H.B. Kagan, Compt. Rend . Ser. , 1973, 277, 41 . 168. M. Mejstrikova, R. Rericha and M. Kraus, Coll. Czechoslov^-Chem. Comm^. 1974, 39, 135.

169. L.D. Rollmann, J. Am. Chem. Soc.r 1975, 97, 2132. 170. J.P. Collman and C.A. Reed, J. Am. Chem. Soc..

1973, 95, 2048. 194

171. T.A.M.M. Maas, M. Kuijer and J. Zwart, J. Chem. Soc. Chem. Comm.r 1976, 86. 172. R.J. Card and D.C. Neckers, Inorc. Chem.f 1978,

17, 2345. 173. C.U. Pittman, Jr., P.L. Gruke, O.E. Ayers, S.P. McManns, M.D. Rausch and G.A. Moser, J . ,Polm.-SsJL...Polym. Chem«...EtU, 1972, 10,

379. m . J. Manassen, Israel_J. Chem.T 1970, 8, 5. 175. S.V. McKinley and J.W. Rakshys (Dow Chemical Co.), U.S. Pat. 3 708 462. 176. G.L. Baker, S.J. Fritschel and J.K. Stille, J. Org. C h e m . 1981 , 46, 2960.

177. H.C. Meinders and G. Challa, J. Mol. Catal.T 1980, 7, 321. 178. W. Bre'emhaar, H.C. Meinders and G. Challa, J. Mol. Catal...f 1981, 10, 33.

179. E. Tsuchida, M. Kanenko, H. Nishide and Y. Kurlmura, Mokromal. Chem.. 1974, 175, 171. 180. P. Bedetti, U. Carunchio and E. Cernia, J. Polvm. Sci^ Polvm. Letts^_Ed^. 1973, 13,

329. 181. M. Furue, K. Sumi, and S.I. Nozakura, Chem. Lett.. 1981, 1349.

182. T. Ikeda, C.R. Leidner and R.W. Murray, J. Am. ChemJ_Soc.r 1981, 103, 7422. 195

183. J.P. Collman and C.A. Reed, J. Am. Chem. Soc. t 1973, 95, 2048. 184. E. Tsuchlda and H. Nlshlkawa, J. Polym^Scl^ Polvm. Chem. Ed., 1976, 14, 1557. 185. Y. Inakl, K. Kimura and K. Takemoto, Makromol. Chem.T 1973, 171, 19. 186. D.F. Herman, U.S. Pat. 2940957 (1960).

187. V.V. Korshak, A.M. Sladkov, L.K. Luneva and A.S. Girshovich, Vvsokomol. Soed., 1963, 5, T284. 188. I.A. Alekseeva, 0.1. Abramova and A.L. Suvorov, Yy.SQkpm,Ql, .Soed.,., 1976, 18B, 31. 189. N. Fujita and K. Sonogashira, J. Polvmer. Scl.r P.olypifir-.Chfiin...EriL,., 1974, 12, 2845. 190. G.G. Skvortsova, E.S. Domnina, Yu. N. Ivlev and N.H.__ Chipanlna, ZZhur_Qbshoh. Khlm., 1976, 46,

1593. 191. C.E. Carraher, J.E. Sheats and C.U. Pittman, Jr., J....Macromol. Sol., A, 1981, 6, 1. 192. C.U. Pittman, Jr., in nMetalloorganicheskie Polimerov" (Organometalllc Polymers), Izd. Mir, Moscow, 1981, p. 9. 193. C.U. Pittman, Jr., R.L. Voyes and J. Elder, Macromoleculea, 1971, 4, 302.

194. K. Hamann and R. Laible, F.A.T.I.P.E.C. Jahrbuch, 1978, 14, 17. 195. A.J. Schouten, Thesis, Groningen, 1980. 196

196. R.H. Grubbs and L.C. Kroll, Polvm. Preprints, /Am. Chem. Soc. Dlv. Polvm. Chem.. 1971, 12,

423. 197. R.H. Grubbs and L.C. Kroll, J. Am. Chem. Soe.t

1971, 93, 3062. 198. R.H. Grubbs, L.C. Kroll and E.M. Sweet, Polvm. Preprints. Am. Chem. Soo. Dlv. Polvm._Chem..

1972, 13, 828. 199. R.H. Grubbs, E.M. Sweet and S. Phisanbut, in "Catalysis In Organic Synthesis" ed. P. Rylander and H. Greenfield, Academic Press, New York, 1976,

p. 153. 200. C.U. Pittman, Jr. and O.Y. Ng, U.S. Pat. 4 258 206

(1981). 201. C.U. Pittman, Jr., L.R. Smith and R.H. Hanes, J. Am. Chem. Soc^. 1975, 97, 1742. 202. E. Tsuchida and H. Nishide, IUPAC Symp on Macromolecules, Prepr., Florence, 1980, Vol. 4, p.

147. 203. W.O. Haag and D.D. Whitehurst, in "Catalysis", ed.

J.W. Hightower, 1973, Vol. 1, paper 29, p. 2165. 204. W.D. Bonds, Jr., C.H. Brubaker, Jr., E.S. Chandrasekaran, C. Gibbons, R.H. Grubbs and LL.C. Kroll, J. Am. Chem. Soo.T 1975, 97, 2128. 205. R.K. Kochkar and R. Petit, J. Oreanomet. Chem..

1966, 6, 272. 197

206. G. Gubitosa, M. Boldt and H.H. Brintzinger, J. Am. Chem. Soc.. 1977, 99, 5174. 207. S.E. Jacobson, W. Clements, H. Hiramoto and C.U. Pittman, Jr., J. Mol. Catal.. 1975, 1, 73. 208. C.U. Pittman, Jr. and L.R. Smith,

J-I- Am 1..CJiegl-. Sgg.i., 1975, 97, 1749. 209. C.U. Pittman, Jr. and L.R. Smith, J. Am. Chem. Soc.f 1975, 97, 341. 210. T. Kauffman, J. Koenig, A. Walterman, Chem. Ber.T 1976, 109, 3864. 211. E. Buhleier, W. Wehner and F. Vogtle, Cilfiffl. 1978, 111, 200. 212. N.H. Prizada, P.M. Pojer and L.A. Summers, Z. Naturforsch^. 1976, 316, 115. 213. S.P. Massie, Iowa State Coll. J. Sci.. 1946,

21, 41. 214. T.M. Bare and H.O. House, Org. Syntheses, 49,, 81. 215. P.E. Spoerri and A. Erickson, J. Am. Chem. Soc.. 193860, 400. 216. G. Black, E. Depp and B.B. Corson, J. Chem. Soc.. 1948, 14.

217. R.F. Larock, J. Ore. Chem., 1974, 25, 3721. 218. E. 0. Chiai, J. Org. Chem.P 1953, 18, 534. 219. Y. Sprinzak, Org. Svn. Coll. Vol. IV. 91. 220. R.F. Heck, Org. Reaction. 1982, 27, 345. 198

221 . A. Musco, V. Kuran, A. Silvani and M.W. Anker, J.. Chem..Sp.c. Chem..C.gjnmurL.., 1973, 938. m. Matsumoto, H. Yoshioka, H. Yoshloka, K. Nakatsu, T. Yoshlda and S. Otsuka, J . Am. Chem. Soc_. , 1974, 96, 3322. 222. J.R. Durig, R. Layton, D.W. Sink and 6.R. Mitchell, Spectrochlmlca Aota. 1965, 21, 1367; J.R. Durig, B.R. Mitchell, D.W. Sink, J.N. Willis, Jr. and A.S. Wilson, Spectrochlmica Jlcta. 1967, 23A, 1121.

223. A. Spencer, J. Organomet. Chem.t 1983, 258, 101. 224. P.A. Lay, Xnorgi ..Chenu. 1984, 23, 4775.

225. H. Gusten and L. Klasinic, Tetrahedron, 1968, 24, 5499. 226. C.S. Wood and F.B, Mallory, J. Ore. Chem., 1964,

29, 3373.

227. O.H. Wheeler and H.N.B. de PaBon, J. Org. Chem,., 1965, 30, 1473. 228. H. Gusten and M. Salzwedel, Tetrahedron, 1967,

23, 173. 229. H.G. Neumann, Hoppe-Seyler's Z. Physiol. Chem., 1967, 348, 313. 230. J.M.J. Frechet, J. Warnock and M.J. Farrall, J. Qrg. Chenu, 1978, 43, 2618. 231. R. Belcher and L. Gordon, nThe Analytical 199

Chemistry of the Nobel Metals", 1966, 24, 434. 232. C.J. Cavallito, P.A. Gray, U.S. Pat. 3,409,626. 05 Nov. 1968. 233. T . A ‘. Stephenson, S.M. Morehouse, A.R. Powell, J.P. Heffer and G. Wilkinson, J. Chem. Soc., 1965, 3632. 234 . J.M. Snell and A. Welssberger, Ore. Svn. Coll...

y.Pl. .Ill, 788. 235. F.T. Faug, et al., J. Am. Chem. Soc.. 1958,

80, 563. 236. C.R. Conard and M.A. Dolliver, Org. Svn. Coll.

Vol. II. 167. 237. T. Ukai, H. Kawazura and Y. Ishii, J . Orcanomet.

QhSSUt 1974, 65, 253. 238. L. Eberson and L. Johsson, Acta. Chem. Scand., B30, 1976, 579. 239. S. Laskin, M. Kuschner, R.T. Drew, V.P. Capplello

and M. Nelson, Ar.cb....En.Y.ir.PiL.-Hsal^h, 1971, 23, 135; L.D. Taylor and N.S. Simon, J. Phvs. Chem.. 1974, 78, 2696. 240 . A. Warshawsky and N. Shoef, J. Polvm. Sci.f Polvm. Chem. Ed.. 1985, 23, 1843. 241. M. Camps, M. Chatzopoulos, J.P. Montheard, Makromol. Chem. Rapid Commune. 1982, 3, 35. 242. S. Tanimoto, T. Miyake, M. Okano, Svnth..Commun^. 1974, 4, 193. 200

243. 0. Grummitt and A. Buck, Org. Svnth. Coll.

195. 244. C.H. Banford and H. Lindsay, Polymer. 1973,

14, 330. 245. J.P. Montheard, M. Camps, C.R. Seances Acad.

Sci.r SerC.. 1980, 291, 117.

246. C.L. Arcus and N.S. Salomons, J. Chem. Soc.^,

1962, 1515. 247. R. Arshady, G.W. Kenner and A. Ledwith,

Hakromol. Chem.. 1976, 177, 2911. 248. T. Satake, K. Maruyama, Japan. Kokai 73,103,528,

25 Dec. 1973.

249. W.J. Dale, L. Starr and C.W. Strobel,

J. Org. Chem.. 26, 2225. 250. R.H. Wiley and H.H. Patrick, J. Am. Chem. Soc..

1949, 71, 2429. APPENDIX Attempted synthesis of p-vinylbenzyl chloride: Chloromethylated polystyrene affords a convenient route for the modification of polystyrene with a wide variety of functional groups, such as ester, amine, alkyl, quaternary ammonium salt, phosphonium salt and so on. Many diverse applications using polymers as supports are then accessible such as phase-transfer catalysts, hydrogenation catalysts, mordants in photography, and membrane-modified electrodes. The preparation of chloromethylated polystyrene is, thus an intriguing topic. Direct chloromethylation of polystyrene beads with chloromethyl methyl ester and bis-chloromethyl methyl ether proves to be the most convenient and popular method employed. However, there are some limitations to this procedure. Ca) The essential reagent, chloromethyl methyl ether, is highly carcinogenic, (b) Isomers (p- and m-) of chloromethylated aryl were generated kinetically. (c) Preformed chloromethylated polystyrene does not react selectivity with bifunctional reagents, (d) the by-products formed are retained on the polymer support. The problems associated with the later three disadvantages can be eliminated by functionalization of

201 202

pure p-vinylbenzyl chloride, purification of the "modified" monomer and then polymerization to produce the "modified" polystyrene. 240 Recently A. Warshawsky and N. Shoef developed a polymer-bound chloromethylating agent with much lower toxicity than low molecular weight analogues. They are specially useful in the synthesis of pharmaceuticals when a safe halomethylation of aromatic substrates produces the required intermediates is required. However, these polymeric halomethylating regents are too expensive to consider for commodity synthesis and are moisture-sensitive. Unfortunately, there are no direct available sources for pure p-vinylbenzyl chloride. The commercially available material is in fact, a mixture of meta (“ 60%) and para (" 40%) isomers. J.P. Honth’eard claimed a successful separation of p- and m- isomers via 241 a hydrogen brominated Intermediate. The mixture was subjected to an antiMarkonikov addition of HBr (HBr/AIBN), the pure p-( ft -bromoethyl)benzyl chloride was isolated and converted into p-vinylbenzyl chloride by dehydrobromination with sodium t-butoxide (equation 54). 203

HBr, AIBN . t-BuoNa ch2ci Br

174 p- and m-

ch2ci (54)

173

p-( ft -BromoethyDbenzyl chloride was also synthesized by Friedel-Crafts alkylation of p -phenylethyl bromide in only 292 yield with chloromethyl methyl ester catalyzed by A1C1, in Cl-pCHCHClp- 2H2 c In our modification of this synthesis, freshly prepared chloromethyl methyl ether/methyl acetate solution and SnCl^ were used instead of using AlCl^, pure chloromethyl methyl ether and ClgCHgCHgCl'g• An improved yield of 34.3? of chloromethylated p-isomer was isolated. However, this is not our final goal since the utilization of carccinogenic chloromethyl methyl ether must be avoided. Although 1-chloromethylnapthalene was synthesized from the reaction of naphthallene and paraformaldehyde catalyzed by concentrated hydrochloric acid in 7M~77% yield, J a similar reaction was attempted on (2-bromoethyl)benzene without success. (2-bromoethyl) benzene was, in another trial, treated with 2 0 4 para-formaldehyde and HC1 gas in the presence of a Lewis acid ZnCl^ in CS2 to afford less than 10% yield based upon an NMR analysis on the reaction mixture. An increase of HCl(g) flow rate and extension of reaction time to two days failed to improve the yield. Bamford claimed a successful preparation of p-vinylbenzyl chloride simply by direct chloromethylation of styrene with concentrated HC1 and ohh aqueous formaldehyde. After several attempts to repeat Bamford's procedure, it was apparant that it was impossible to make the required p-isomer by this simple method. Several known procedures for the preparation of p-vinylbenzyl chloride from the corresponding para-precursors were summarized from equations 55 to 59. The most intriguing synthesis of p-vinylbenzyl chloride under laboratory conditions is the approach starting from p-formylstyrene. As shown in equations 60 and 61, p-formyl styrene was synthesized from the reaction of p-vinylphenylmagnesium chloride and dry dimethylformamide or from the condensation of phthalaldehyde with malonic acid followed by decarboxylation of p-formylcinnamic acid with quinoline/Cu. 205

PPh3-CClA |A) -LiA1H4> O - Q (55) CHO c h 2o h c h 2ci 170 169 173 HO dehydration NaBH, o (56) CH Cl CH2C1 173

,C1 * o CH,MgBr3 CN CHO ON 177 176 179 ;h 2ci

O (57)

HO 160 173

Cl Na0H/CH,0H S0-C1,,, hv O K0H/(H0CH,,CH,),0

181 182 183

OCH. BC1. (5B) Q CClA /0*c O

c h 2o c h 3 CH2C1 184 165 173

furnace tube (59) O. 200-700 c, Cl2 * O CH2C1 182 173 206

However, both approaches do not appear attractive owing to the polymer formation (equation 60), and tedious Soxhlet extraction and high temperature decarboxylation (equation 61) respectively.

Mg/THF DMF (60)

170

c o 2h | W | CH2(C0gH)2 ^ r W (6 1) pyridine, ^ ' quinoline,A ethanol CHO 0 CHO 166 187 170

Terephthaldehyde, synthesized from photobromination of p-xylene followed by hydrolysis, was employed as precursor for the preparation of p-formylstyrene. As shown in Scheme XXX, in an attempt to prepare p-formylbenzylalcohol (167) by reduction of terephthaldehyde with 1/M NaBH^, an inseparable mixture of 166, 167 and 168 in the molar ratio of 207

SCHEME XXX Attempted preparation of p-vinylbenzyl chloride from terephthaldehyde OH CHO CH-.OH 4 NaBH4 o ethanol - > 0 + O CHO CHO CHpOH 166 166 16B 38.07% 35-24% 2 6 .68% PPh3=CH2

PPh?=CH2

CH,

* V CHO 169 173 1J70 + j jcH2=CH? 6 ! |Na2PdCl4 CHnOH -> known or done ■* planned, <> not done Br ^ not 172 working future goal NaBH.

0 208

35.24?:38.07? : 2 6 . 6 8 ? was obtained. In another attempt, 166 was treated with a Wittig reagent (Ph^PCH^Br/n-BuLi/ether) to produce a milky

suspension, no p-formylstyrene, 170, was isolated. The failure was probably due to the poor solubility of 166 in diethyl ether. p-Bromobenzaldehyde, 171» was successfully reduced to p-bromobenzylalcohol, 172, in good yield. No further try has been carried out on the vinylation of 172 with olefin has been carried out on the vinylation of 172 with ethylene gas catalyzed by Na2PdCli(. This route, however, is a very promising one for the preparation of 169 which then can be converted to p-vinylbenzyl chloride 173 by a known reaction. The Experimental procedures mentioned in this appendix has been included in Chapter V. 209

VITA

Chia-Hsing Sun was born May 26, 1954, in Taiwan, Republic of China. He attended public schools in Taipei and graduated from Chen-ko High School in 1972. At that year he entered Tunghai University in Taichung and a B.S. in Chemistry was awarded in 1976. He then served in the army for two years. In 1978, he joined the graduate program in Tsinhua University under the instruction of Dr. Deiryush Cheng. A M.S. in Chemistry was awarded in 1980. At that summer (August 8, 1980), he arrived in the U.S.A. and became a graduate student at Louisiana State University in Baton Rouge. From 1980-1985 he was employed as a Teaching Assistant in Chemistry. Currently he is a candidate for the Doctor of Philosophy degree with a major in Organic Chemistry and minor in Inorganic Chemistry. He is going back to his country and contribute what he has learned to the society.

t DOCTORAL EXAMINATION AND DISSERTATION REPORT

Candidate: Chia-Hsing Sun

Major Field: Organic Chemistry

Title of Dissertation: Palladium Catalyzed Vinylic Substitution Of Aryl Halides On Polymeric Nitrogen Supports

Approved:

Major Professor and Chai

Dean of the Graduate School

EXAMINING COMMITTEE:

Jz&m I

Date of Examination:

November 21, 1985